Artwork for podcast Learning Bayesian Statistics
#95 Unraveling Cosmic Mysteries, with Valerie Domcke
Episode 95 • 15th November 2023 • Learning Bayesian Statistics • Alexandre Andorra
00:00:00 01:00:16

Share Episode

Shownotes

Proudly sponsored by PyMC Labs, the Bayesian Consultancy. Book a call, or get in touch!

Welcome to another installment of our LBS physics deep dive! After exploring the world of experimental physics at CERN in our first video documentary in episode 93, we’ll stay in Geneva for this one, but this time we’ll dive into theoretical physics.

We’ll explore mysterious components of the universe, like dark matter and dark energy. We’ll also see how the study of gravity intersects with the study of particle physics, especially when considering black holes and the early universe. Even crazier, we’ll see that there are actual experiments and observational projects going on to answer these fundamental questions!

Our guide for this episode is Valerie Domcke, permanent research staff member at CERN, who did her PhD in Hamburg, Germany, and postdocs in Trieste and Paris.

When she’s not trying to decipher the mysteries of the universe, Valerie can be found on boats, as she’s a big sailing fan.

Our theme music is « Good Bayesian », by Baba Brinkman (feat MC Lars and Mega Ran). Check out his awesome work at https://bababrinkman.com/ !

Thank you to my Patrons for making this episode possible!

Yusuke Saito, Avi Bryant, Ero Carrera, Giuliano Cruz, Tim Gasser, James Wade, Tradd Salvo, William Benton, James Ahloy, Robin Taylor,, Chad Scherrer, Zwelithini Tunyiswa, Bertrand Wilden, James Thompson, Stephen Oates, Gian Luca Di Tanna, Jack Wells, Matthew Maldonado, Ian Costley, Ally Salim, Larry Gill, Ian Moran, Paul Oreto, Colin Caprani, Colin Carroll, Nathaniel Burbank, Michael Osthege, Rémi Louf, Clive Edelsten, Henri Wallen, Hugo Botha, Vinh Nguyen, Marcin Elantkowski, Adam C. Smith, Will Kurt, Andrew Moskowitz, Hector Munoz, Marco Gorelli, Simon Kessell, Bradley Rode, Patrick Kelley, Rick Anderson, Casper de Bruin, Philippe Labonde, Michael Hankin, Cameron Smith, Tomáš Frýda, Ryan Wesslen, Andreas Netti, Riley King, Yoshiyuki Hamajima, Sven De Maeyer, Michael DeCrescenzo, Fergal M, Mason Yahr, Naoya Kanai, Steven Rowland, Aubrey Clayton, Jeannine Sue, Omri Har Shemesh, Scott Anthony Robson, Robert Yolken, Or Duek, Pavel Dusek, Paul Cox, Andreas Kröpelin, Raphaël R, Nicolas Rode, Gabriel Stechschulte, Arkady, Kurt TeKolste, Gergely Juhasz, Marcus Nölke, Maggi Mackintosh, Grant Pezzolesi, Avram Aelony, Joshua Meehl, Javier Sabio, Kristian Higgins, Alex Jones, Gregorio Aguilar, Matt Rosinski, Bart Trudeau, Luis Fonseca, Dante Gates, Matt Niccolls and Maksim Kuznecov.

Visit https://www.patreon.com/learnbayesstats to unlock exclusive Bayesian swag ;)

Links from the show:

Abstract

by Christoph Bamberg

Episode 95 is another instalment of our Deep Dive into Physics series. And this time we move away from the empirical side of this topic towards more theoretical questions. 

There is no one better for this topic than Dr. Valerie Domcke. Valerie is the second researcher from the CERN we have on our show. She is located at the Department of Theoretical Physics there.

We mainly focus on the Standard Model of Physics, where it fails to explain observations, what proposals are discussed to update or replace it and what kind of evidence would be needed to make such a decision.

Valerie is particularly interested in situations in which the Standard Model brakes down, such as when trying to explain the excess gravitational pull observed that cannot be accounted for by visible stars. 

Of course, we cover fascinating topics like dark matter, dark energy, black holes and gravitational waves that are places to look for evidence against the Standard Model.

Looking more at the practical side of things, we discuss the challenges in disentangling signal from noise, especially in such complex fields as astro- and quantum-physics. 

We also touch upon the challenges Valerie is currently tackling in working on a new observatory for gravitational waves, the Laser Interferometer Space Antenna, LISA. 

Transcript

This is an automatic transcript and may therefore contain errors. Please get in touch if you're willing to correct them.

Transcripts

Speaker:

Welcome to another installment of our LBS

physics deep dive.

2

:

After exploring the world of experimental

physics at CERN in our first video

3

:

documentary in episode 93, we'll stay in

Geneva for this one, but this time we'll

4

:

dive into theoretical physics.

5

:

We'll explore mysterious components of the

universe, like dark matter and dark

6

:

energy.

7

:

We'll also see how the study of gravity

intersects.

8

:

with the study of particle physics,

especially when considering black holes

9

:

and the early universe.

10

:

Even crazier, we'll see that there are

actual experiments and observational

11

:

projects going on to answer these

fundamental questions.

12

:

Our guide for this episode is Valérie

Dormcke, permanent research staff member

13

:

at CERN who did her PhD in Hamburg,

Germany, and postdocs in Trieste and

14

:

Paris.

15

:

When she's not trying to decipher the

mysteries of the universe, Valérie can be

16

:

found on

17

:

she's a big sailing fan.

18

:

This is Learning Vagin Statistics, episode

,:

19

:

Hello my dear Vagins!

20

:

Some of you have reached out for advice

and coaching in parallel to my online

21

:

courses on intuitivevagin.com.

22

:

So, to help you, I have started something

new.

23

:

If you go to

24

:

You can pair your online course with my

15-hour or 20-hour coaching packages to

25

:

get a fully premium learning path.

26

:

Each week, we'll get on a one-to-one call

and we'll walk through any questions,

27

:

difficulties, or roadblocks that you may

have to jumpstart your learning even more.

28

:

Again, that's topmate.io slash Alex

underscore and Dora.

29

:

And now, let's talk theoretical physics

with Valerie Donka.

30

:

I'll show you how to be a good peasy and

change your predictions.

31

:

Valérie Damke, welcome to Learning Asian

Statistics.

32

:

Glad to be here.

33

:

Yeah.

34

:

Thank you for taking the time.

35

:

I am really happy to have you on the show.

36

:

Again, a physics-packed episode.

37

:

I'm really, really happy about that and I

have a lot of questions for you.

38

:

I think you're the first theoretical

physicist to come on the show.

39

:

That's cool.

40

:

We're going to talk about topics.

41

:

a bit different than those we talk about

when we have experimental physicists on

42

:

the show.

43

:

So that's cool, more diversity for the

listeners.

44

:

And also, when that episode is going to

air, by the magic of time travel, episode

45

:

93 will have been published.

46

:

So that's the one at CERN.

47

:

So the very special video documentary I

did at CERN with Kevin Kaif.

48

:

So if listeners haven't checked it out

yet, I highly recommend it.

49

:

And that one, of course, I recommend

mainly watching the YouTube video because

50

:

I recorded and edited it firstly for video

format.

51

:

You have access to the audio format also,

but I'm telling you, it's going to be

52

:

funnier in video.

53

:

So now to actually complete what we talked

about in episode 93.

54

:

where Kevin does a lot of fun experiments

at CERN, today we are going to talk about

55

:

another part of physics that's done at

CERN, thanks to you, Valerie.

56

:

But first, before doing that, let's start

with your origin story.

57

:

How did you come to the world of

theoretical physics, and how sinuous of a

58

:

path was it?

59

:

It was, it was more of a path that I kind

of ended up on without honestly thinking

60

:

about it too much.

61

:

It's kind of been a topic that has

fascinated me since I was quite young,

62

:

reading science fiction books and the

like.

63

:

And I basically, we just kind of following

my interests, taking the course of the

64

:

university that interests me most without

thinking too much about where that would

65

:

lead me in the end.

66

:

And it was basically only when I was doing

my PhD that I realized, wow, I'm actually

67

:

working on cosmology and kind of these big

open questions of the universe, which is

68

:

something I was dreaming about as a kid.

69

:

And somehow I got there without, somehow

without too much planning, but just

70

:

following what I thought was kind of the

most interesting thing for me to do at

71

:

every step.

72

:

Oh yeah, so it's really like the call of

passion for you.

73

:

In a sense, in a sense.

74

:

Yeah, that's really cool.

75

:

I mean, and that's also one of the cool

things of this kind of job, right?

76

:

In physics or I don't know, airplane

pilots or firefighters.

77

:

You can dream about them already as you're

a kid and then make that your job.

78

:

I personally love my job, but...

79

:

I'm afraid I cannot say that I dreamed

about patient statistics when I was a kid.

80

:

Like I never told when I was five years

old, oh, I want to be a patient

81

:

statistician.

82

:

You know, that's not how it works,

unfortunately.

83

:

Really?

84

:

Yeah, no, I know, I know that must be

quite disappointing to a lot of people,

85

:

but I had to burst that bubble because I

get a lot of questions about that, yeah.

86

:

So.

87

:

I would also say that you kind of have to

really...

88

:

dream about or be enthusiastic about it,

because doing science, you always

89

:

encounter moments when nothing works.

90

:

Yeah.

91

:

And if you're not passionate about

actually solving the problem, it's, you're

92

:

just going to get stuck.

93

:

Yeah.

94

:

No, definitely.

95

:

That's a very good point.

96

:

And that's where actually statistics get

back in the, in the mix, because that's, I

97

:

would say that's the same for

98

:

programming and the kind of statistics at

least I do where you are going to get a

99

:

lot of bumps along the way.

100

:

And I always say to beginners that models

never work, only the last iteration of a

101

:

model is going to work.

102

:

And even then, you just have to be

satisfied with good enough.

103

:

So that's a field where you have to become

comfortable failing all the time.

104

:

First, be comfortable with making mistakes

and failing.

105

:

And also where you need to be driven by

passion because if you don't have that

106

:

inherent passion, you're not going to

still be driven to solve those numerous

107

:

data analysis issues and bugs and stuff

like that.

108

:

So now, I'd like to talk about what you do

actually, what you're doing nowadays,

109

:

because we know you dreamt about doing

that since you were a child.

110

:

But how would you define the work you're

doing nowadays and what are the topics

111

:

that you are particularly interested in?

112

:

Right, I think there's probably two parts

to that question, right?

113

:

One is kind of how does an everyday day

actually look like?

114

:

And the other one is, okay, what are the

big topics I'm interested in?

115

:

Yeah.

116

:

So to start with the format, so what my

day does not look like is that I kind of

117

:

sit in my office all by myself, waiting

for the fantastic idea that is going to

118

:

win me a Nobel Prize.

119

:

That's kind of the image I had maybe as a

kid of how a theoretical thesis would

120

:

work.

121

:

But that's not at all what my day looks

like.

122

:

Right.

123

:

So I'm it's a lot discussing with people,

listening to talks, going to conferences,

124

:

reading papers, discussing over coffee on

a blackboard over lunch.

125

:

And then progress comes bit by bit.

126

:

But it does kind of, there's never a lack

of things to work on.

127

:

There's never a lack of interesting

questions.

128

:

There's only always a lack of time to

decide what is the most interesting

129

:

question of all the questions to work on.

130

:

Because there's really a lot of things

that we don't understand.

131

:

And that brings me a bit to the

overarching team of my research.

132

:

So I work on the intersection of particle

physics and cosmology.

133

:

So

134

:

meaning kind of the physics of the very,

very smallest particles, the fundamental

135

:

building blocks of nature.

136

:

And at the same time, the physics of the

very largest scales, so the largest scales

137

:

we can observe in our universe, and how

the latter can teach us something about

138

:

the former.

139

:

So how kind of from astrophysical or

cosmological observations, we can learn

140

:

something about what is really the nature

of the fundamental building blocks of

141

:

nature.

142

:

Yeah, so small topics, fundamental

building blocks of nature.

143

:

Yeah, thanks.

144

:

That's interesting.

145

:

I'm actually curious.

146

:

So of course, we're going to talk about

the projects you work on a day to day a

147

:

bit more.

148

:

But also I'm curious now that you brought

up basically what your days look like

149

:

concretely.

150

:

Yeah.

151

:

What's the part of basically solitary work

with pen and paper?

152

:

What's the proportion of that in

comparison to, as you were saying,

153

:

collaboration with people, exchange of

ideas and things like that?

154

:

Because I think when you tell people

you're a theoretical physicist, and that's

155

:

definitely the case when you tell people

you're a statistician, most of the people

156

:

doing math on a blackboard.

157

:

So most of the time, which is not true if

you're a statistician.

158

:

So yeah, I'm curious how it is on your

slide.

159

:

Yeah, it's probably similar.

160

:

I mean, if I get one or two hours on block

to actually sit down and do a calculation,

161

:

that's rather the exception than the rule.

162

:

So it is, of course, part of my job, and I

enjoy it a lot.

163

:

Sometimes just to have time just to think.

164

:

really thoroughly about a problem, either

analytically, so pen and paper, or coding.

165

:

But it's usually not like very long

stretches at a time because then you

166

:

either you hit a problem, right?

167

:

Or you hit a solution.

168

:

And in either case, that's the point to

reach out to your collaborators and

169

:

discuss the next steps.

170

:

Yeah.

171

:

I mean, that's interesting because for me,

now I'm using more and more the

172

:

excuse of teaching to dive deep in a topic

and a project because, well, I have to be

173

:

able to explain it properly to students.

174

:

So that's actually, these are actually the

good occasions and rare, quite rare

175

:

occasions where I can just be myself

working on the computer or sometimes with

176

:

a pen and paper and really understand

deeply.

177

:

a topic that I need and want to understand

because otherwise, yeah, you have so many

178

:

other projects and solicitations that can

be hard to actually take the time just for

179

:

yourself and focus on these.

180

:

So I'm the same.

181

:

I do appreciate these solitary moments,

although I'm happy that they are not 90%

182

:

of the work, I have to say.

183

:

Yeah, same here.

184

:

And actually, Sue, you...

185

:

You're a very math savvy person.

186

:

So of course you know about patient stats,

but I'm curious if you were introduced to

187

:

Bayesian methods actually, you know, in

your graduate studies or before, and if

188

:

you use them from time to time in your own

work.

189

:

No, so I never received any.

190

:

any type of formal or informal training.

191

:

So it's, of course, it's something we need

to know in the sense that we deal with

192

:

empirical data.

193

:

Even if I myself don't usually deal

directly with the empirical data, but I

194

:

kind of deal with the processed empirical

data, or I deal with the publications that

195

:

people have written on the data, and then

I need to evaluate, interpret, and kind of

196

:

continue to work from there.

197

:

But for that, of course, I need to kind of

understand the significance of certain

198

:

experimental results.

199

:

So I would say, okay, I mean, I have a

fundamental understanding of them, right.

200

:

But it's, it's not something that actually

kind of on a on a day to day basis, I

201

:

really am like deep in the in the details

of it.

202

:

Yeah, because I'm more work at the kind of

one level.

203

:

away, right?

204

:

So kind of that I that I kind of take, I

need to understand what is the

205

:

significance of that result, right?

206

:

But once I've understood that, I can

basically work directly with the result

207

:

without having going to back to the data

at every step.

208

:

Which is quite a luxury, I have to say.

209

:

I'm a bit jealous.

210

:

I'm very, very happy that there's people

who do the work that I don't need to do.

211

:

Yeah, that's, that's a very good point.

212

:

I like that.

213

:

And if you go listen to episode 93, you'll

see the difference between basically that

214

:

kind of work that Valery does and the

experimental physics work where statistics

215

:

is way more present and of course, patient

statistics is extremely helpful.

216

:

So I find that super interesting to

notice.

217

:

Just because you don't use patient stats,

Valery doesn't mean that your work is not

218

:

interesting.

219

:

I have to put that out there.

220

:

On the contrary, I find it fascinating.

221

:

So let's dive in because one of your areas

of interest is to go beyond the standard

222

:

model phenomenology to kind of probe it,

if I understood correctly.

223

:

So can you tell us what that means and

maybe first define the standard model for

224

:

us?

225

:

Right.

226

:

So the standard model basically reflects

our current understanding of these

227

:

fundamental building blocks of nature.

228

:

So it kind of contains what we think are

kind of elementary particles, which are no

229

:

longer further dividable into even smaller

particles.

230

:

And there's not many of them.

231

:

There's basically a handful of them,

depending how you count.

232

:

And we think that these...

233

:

fundamental particles together with the

interactions between these particles that

234

:

they explain all of kind of nature, the

way it surrounds us, right?

235

:

So all, all everything that we can, we can

grasp, grasp or experience here on earth.

236

:

And the standard model describes basically

this.

237

:

So it describes kind of which building

blocks are there and how do they interact

238

:

with each other.

239

:

And now going beyond the standard model,

because a model is always a model, right?

240

:

So it means that it describes kind of

nature to the best of our knowledge.

241

:

But most models are incomplete at some

level, right?

242

:

Because because it's kind of only a way

that we describe nature, not actually the

243

:

fundamental theory of nature.

244

:

And for this, the standard model of

particle physics, in particular, it, it

245

:

does extremely well in many respects, one

could even say,

246

:

frustratingly well, because like in all

our searches of looking for new

247

:

interactions, looking for new particles

here at CERN at the Large Hadron Collider,

248

:

we always keep confirming the predictions

that the standard model makes with

249

:

incredible accuracy.

250

:

But we still know the model is not

complete.

251

:

And the reason we know that the model is

not complete basically comes from

252

:

cosmology.

253

:

So there's observations that we make about

the dynamics of the universe.

254

:

or properties of the universe, which are

simply in contradiction with this model,

255

:

which tells us that there's ingredients

missing.

256

:

And we have a rough idea of what these

ingredients are.

257

:

Or rather, maybe, instead of one rough

idea, we have 100 rough ideas.

258

:

And the big question is, which one of

these is correct?

259

:

Is any one of these correct?

260

:

And how can we make progress in

understanding these missing parts better?

261

:

So to give you some keywords, things like

dark energy, dark matter, those are some

262

:

of the open questions.

263

:

Yeah, because we know basically you say

they are open because first we cannot

264

:

really explain them fully for now, as we

said in episode 93, but also we know that

265

:

the standard model breaks down at those

points and cannot explain them.

266

:

So that's basically what you're trying to

understand.

267

:

Why does the standard model fail here and

how can we actually explain these

268

:

phenomena?

269

:

Correct.

270

:

I see.

271

:

So concretely, what does that research

look like?

272

:

Maybe could you share an example of a

discovery or theoretical development in

273

:

this field that has the potential to

reshape our understanding of particle

274

:

physics?

275

:

You mean like a discovery in the past that

did that or a discovery, potential

276

:

discovery in the future that...

277

:

I would say both.

278

:

Yeah, both if you can.

279

:

Let's start with the past.

280

:

So one observation, for example, was

rotation curves of galaxies.

281

:

So people were looking at galaxies in the

sky.

282

:

And they were they were looking at kind of

how fast the stars were rotating, which

283

:

you can do by measuring the redshift of

the stars.

284

:

Because as they move away from us, the

light gets slightly red as they move

285

:

towards us, the light gets slightly bluer.

286

:

And if you know, like if you have an

object on a stationary orbit, and you

287

:

know, you know, the orbit, you know, the

velocity.

288

:

I mean, actually, even knowing the orbit

and the mass of stars enough.

289

:

Then you can estimate how much mass you

need in a center in order to make that a

290

:

stable orbit.

291

:

And so that's just Newton dynamics, high

school physics.

292

:

And what people observed is that the mass

that you needed in the center in order to

293

:

put these stars on the orbits that were

being observed was much, much bigger than

294

:

the mass you would have inferred just by

counting stars.

295

:

And now you can say, OK, well, counting

stars is obviously not enough, right?

296

:

Because there's going to be planets.

297

:

Planets are not luminous.

298

:

So there's going to be a bit of an offset,

but you would have expected that counting

299

:

stars would give you a good estimate.

300

:

And it turned out it was completely off.

301

:

So it turned out it was kind of a big

amount of something that has an attractive

302

:

gravitational force in the center of the

galaxies, or like in a halo around the

303

:

galaxies, which was invisible to our

telescopes.

304

:

And that is basically what I'm coined the

term dark matter.

305

:

because it kind of has a gravitational

pull of matter, just like everything else.

306

:

But it's dark, meaning we can't see it.

307

:

And not seeing it means like not only kind

of we don't pick it up with telescopes,

308

:

but kind of also all other type of

experiment that we've performed to date,

309

:

trying to find this stuff.

310

:

And this stuff should be around

everywhere, right?

311

:

So it's not that there's none of it on

Earth.

312

:

It's just that it's so incredibly weakly

interacting with...

313

:

Yeah.

314

:

all the instruments that we build, that

it's very difficult to see.

315

:

And then observations, I mean, more

observations, particularly cosmological

316

:

observations, reveal that there's actually

five times more of this dark matter than

317

:

there is of what we call ordinary matter.

318

:

So ordinary matter is everything that we

know of on Earth and everything that we

319

:

can describe with our standard model of

part of the physics.

320

:

Meaning that there's really a lot of stuff

out there that we don't know.

321

:

That's just one example.

322

:

And that kind of gave very clear

indication that the Sonop model of

323

:

particle physics is incomplete.

324

:

And that we're not only missing a little

bit, but that we're actually missing a

325

:

very big bit of the picture.

326

:

And along the same line of thought, you

know, what would really be a

327

:

groundbreaking discovery if one of the

many experiments looking for such a dark

328

:

matter particle, if they would actually

find something.

329

:

I mean, even if they don't find anything,

if a particular experiment doesn't find

330

:

anything, then okay, you still learn

something because you can probably exclude

331

:

some class of models.

332

:

But if one of them actually made a

discovery, and we would have kind of a

333

:

very clear indication of which direction

to go in when we're kind of trying to

334

:

describe these dark matter particles, that

would be a complete game changer.

335

:

Yeah, for sure.

336

:

And so these kinds of experiments are

underway at CERN in particular, right?

337

:

Yeah, at CERN and across the world.

338

:

I mean, it's something you can look for

when in a collider because you can always

339

:

hope that as your collider reaches higher

and higher energy, or you have just more

340

:

and more particles that you're colliding,

you'll eventually kind of reach the

341

:

threshold for producing these particles.

342

:

And then you can find indirect traces of

them.

343

:

in the K channels, or you basically have

some sort of, not a collider, but

344

:

basically just a very big detector volume

somewhere.

345

:

So a very big amount of an noble gas, for

example, even water.

346

:

And then you wait basically for a dark,

you like have to shield it very well

347

:

against everything else.

348

:

And then you wait for some dark matter

particle.

349

:

to have one of these very rare

interactions with one of the atoms of your

350

:

detector.

351

:

And you're looking for that interaction.

352

:

And there's a there's a range of

experiments underway, looking for very

353

:

different types of these dark matter

candidates.

354

:

Yeah, so but we've been we've been hoping

that we'll find it any day now.

355

:

Basically, since I don't know, I mean,

basically, since I do physics.

356

:

So we don't know.

357

:

It could be around the corner or it could

be very well hidden.

358

:

Yeah.

359

:

I mean, these kinds of experiments, I

think I would not be able to work on them

360

:

at least full time, you know, that's

awful.

361

:

Like you're just waiting for something and

you cannot control anything.

362

:

Oh, there's plenty of stuff to do.

363

:

You're not just waiting, right?

364

:

I mean, because you're basically

constantly fighting to reduce noise,

365

:

reduce background, understand noise.

366

:

understand background, argue with somebody

who's making noise in the building next

367

:

door, right?

368

:

And disrupting your experiments.

369

:

So, Yeah, yeah, no, for sure.

370

:

That's, yeah, that's something you have to

deal with all the time, I guess.

371

:

But yeah, I mean, I would be also, you

know, incredibly stressed out.

372

:

Like, so did the, I think a lot of them

are helium pools, right?

373

:

Or something like that.

374

:

Did the helium pool move tonight or not?

375

:

I would be incredibly stressed out.

376

:

Yeah, so thanks a lot.

377

:

That's actually very interesting to hear

about that because I find this kind of

378

:

experiment absolutely fascinating.

379

:

And where does your work come into that

picture?

380

:

So you're part of these big teams, right,

in physics.

381

:

Like you see a physics paper, it's like

most of the time a lot of people, because

382

:

a lot of you are very, like many of you

are very specialized in what they do.

383

:

And so you bring one of the brick to the

paper.

384

:

So you in this kind of work, what do you

do?

385

:

What do you bring?

386

:

So the papers really with like the many

hundreds of authors, they're usually the

387

:

experimental collaborations.

388

:

So.

389

:

As a theorist, you know, I usually have

whatever, two, three, four, co-authors on

390

:

a paper.

391

:

That's a lot.

392

:

Right.

393

:

So we build, of course, very heavily on

the results of these big collaboration

394

:

papers.

395

:

But largely, the work that I concretely do

is with much smaller groups of people.

396

:

So, yeah, I basically have two...

397

:

two main approaches to this.

398

:

One is kind of starting from really

standard model of particle physics, and

399

:

trying to come up with possible extensions

of that, which kind of makes sense within

400

:

the framework that the standard model is

written in.

401

:

So it makes sense within the symmetries

that they are, makes sense within the

402

:

framework of quantum field theory, and

address some of these open problems that

403

:

we have in cosmology.

404

:

And then the question is, okay, once

you've kind of constructed

405

:

such an inherently consistent model, what

sort of implications might that have in

406

:

various types of experiments?

407

:

Right.

408

:

So that can be experiments like the chart

Hadron Collider.

409

:

It can also be some astrophysical

observations, or it can be some

410

:

cosmological observations.

411

:

So that's kind of one approach, and coming

kind of more from the fundamental

412

:

mathematical theory of it.

413

:

My other approach is more the lamppost

approach, meaning, well, you, you look

414

:

where you can look right, and you hope

that nature is kind.

415

:

And they're kind of the my approach is to

say, okay, what types of probes do we have

416

:

of the universe of astrophysical

processes?

417

:

Try and understand as much as possible

about those, and then see what type of

418

:

models or what kind of types of building

blocks of models.

419

:

you could test with these types of

observations.

420

:

And there, for example, the new big player

in the game are gravitational waves.

421

:

Because now since the first discovery with

LIGO and now a tentative discovery in a

422

:

different frequency range this year with

the pulse of timing arrays, that's kind of

423

:

opening up a completely new way of

observing our universe.

424

:

And so there's the potential for...

425

:

for big excitement in that field.

426

:

So I'm also just involved in trying to

understand as much as possible about how

427

:

gravitational waves can reveal something

about the universe.

428

:

Oh, yeah.

429

:

So that's actually fascinating.

430

:

So yeah, talk to us a bit more about that,

basically.

431

:

What can gravitational waves tell us about

the universe?

432

:

And maybe redefine quickly what

gravitational waves

433

:

waves are for listeners?

434

:

Right, so gravitational waves are, we

think of them as perturbations of the

435

:

metric, so perturbations of space-time.

436

:

So the type of gravitational waves that

we've already seen with LIGO and Virgo,

437

:

which are big Michelson interferometers,

so the type of

438

:

which are circling each other and then

finally merging.

439

:

So these are like extremely massive

objects.

440

:

And as you might know, a massive object

kind of creates if you want a dent in

441

:

space-time.

442

:

And if you have two of them, just kind of

their dance around each other really like

443

:

sends out ripples of this kind of

space-time perturbations out into the

444

:

universe.

445

:

If you're very close to a black hole,

right, these ripples will be quite

446

:

significant.

447

:

But then you'd also have all sorts of

other problems, right?

448

:

Because if you're really close to black

hole, I mean, then you have a lot of

449

:

problems.

450

:

So, by the time these gravitational waves

reach us, they've kind of spread out very

451

:

far, meaning the amplitude is very much

decreased.

452

:

So, by the time they reach us, these are

typically very, very small, like tiny

453

:

perturbations in space time.

454

:

So it's not something we have to worry

about in everyday life, rather we need to

455

:

build an extremely sensitive detector to

even pick them up.

456

:

And so, so far, the observations that

we've made are this type of observation.

457

:

So observations of these black holes

merging, which happened, I mean, still at

458

:

the distance of megaparsecs or gigaparsecs

from here, right?

459

:

So it kind of...

460

:

Yeah, quite far away on cosmological

scales.

461

:

But nevertheless, compared to the lifespan

of the universe, these are still fairly

462

:

recent events.

463

:

So at the moment, we're using this to

learn, as a new way to learn about the

464

:

universe surrounding us or the more recent

universe or the relatively recent

465

:

universe.

466

:

Because these gravitational waves are so

weakly interacting with everything, in

467

:

principle, even gravitational waves

generated in the very, very early

468

:

universe, when the universe was not yet

transparent to photons, when kind of no

469

:

other messenger could escape this

primordial soup.

470

:

Gravitational waves could.

471

:

So in principle, if we detected them

today, they could reveal information about

472

:

extremely early times in the universe,

when the temperatures in the universe were

473

:

extremely high, when all the fundamental

particles.

474

:

kind of existed as fundamental particles.

475

:

And when we can really kind of probe these

constituents of the standard model or of

476

:

any model beyond the standard model.

477

:

So that's the ultimate hope.

478

:

But it's challenging because we don't know

what is the amplitude of these gravitation

479

:

waves from the very early universe.

480

:

And so we first need to understand the

gravitation waves generated in the late

481

:

universe.

482

:

Make sure we fully understand that before

we kind of look for a fainter signal.

483

:

Very similar to with photons, right?

484

:

You basically first need to kind of

understand all the light kind of coming

485

:

from the nearby universe, coming from the

galaxy.

486

:

And only when you have a very good

understanding of your foregrounds, can you

487

:

go and can you look for fainter light that

is coming from earlier times.

488

:

Yeah, yeah, that makes sense.

489

:

Because also those waves are like so much

weaker that...

490

:

Also, I'm guessing you have to be a bit

more aware of what you're looking for,

491

:

because otherwise it's even harder.

492

:

And to understand, do we know if...

493

:

Just one black hole, for instance?

494

:

So for instance, the back hole at the

center of our galaxy, is it emitting also

495

:

gravitational waves, but since it's not

orbiting another one, at least that we

496

:

know of,

497

:

the gravitational waves are weaker so we

cannot see them?

498

:

Or do we know that, no, you have to have

the collision of two massive objects to

499

:

get those gravitational waves?

500

:

Yeah, so a single black hole won't do it

because anything that is perfect spherical

501

:

symmetry won't do it.

502

:

That has to do with the fact that these

gravitational waves are tensor modes,

503

:

right?

504

:

So they have two Lorentz indices and

something that's spherical symmetric.

505

:

is a scalar quantity.

506

:

So a single black hole won't do it.

507

:

So you need two, or you need a black hole

and another massive object, so you have a

508

:

black hole and a neutron star.

509

:

Okay.

510

:

Or anything else that breaks spherical

symmetry, right?

511

:

So kind of, I don't know, you dancing

around, right?

512

:

That will in principle generate

gravitational waves.

513

:

They're just very, very small.

514

:

Thank you.

515

:

I'm flattered.

516

:

Yeah, I see.

517

:

Okay.

518

:

Yeah, so it's very like, it's really the

density of the objects that count.

519

:

Yeah, again, you can imagine that.

520

:

A large concentration of mass and in some

asymmetric way.

521

:

So some sort of violent process, which is

condensing a lot of energy, a lot of mass.

522

:

Yeah.

523

:

But in some way that is moving in a bit of

a non-trivial way.

524

:

Yeah, that makes sense.

525

:

Even though I...

526

:

I like thinking about these things because

it's so hard to imagine.

527

:

Like the power of these collisions must be

just incredibly devastating.

528

:

I would love to see that in a way, but

that's so like, it's really impressive and

529

:

at the same time, really frightening.

530

:

Yeah.

531

:

So the, the gravitational waves that we

saw.

532

:

with LIGO, there we think it's something

like two black holes, roughly after the

533

:

mass, like roughly 10 solar masses each

colliding, a bit more.

534

:

And the energy that is just the energy

that is released into gravitational waves

535

:

corresponds roughly to the mass of our

sun.

536

:

So it's a huge amount of energy.

537

:

And now the gravitational waves that we

think we might have seen with these pulsar

538

:

timing arrays.

539

:

These are even more massive objects.

540

:

These are really the large black holes,

right, like the one in the center of a

541

:

galaxy that we think we see colliding.

542

:

So this is two far away galaxies, each

with their big, massive 10 to the 6 solar

543

:

mass black hole in the center.

544

:

And when they collide, that's the signal

that we expect.

545

:

So that's a massive event, right?

546

:

I mean, two galaxies colliding.

547

:

Yeah, you don't want to be close to

witness that.

548

:

Yeah, no, that's for sure.

549

:

These are absolutely fascinating topics

and I'm wondering what are the main

550

:

challenges in understanding these topics

right now and how do you folks as

551

:

researchers in this field...

552

:

address them.

553

:

That's, that's a broad question, right?

554

:

I mean, there's different levels of

challenges, right?

555

:

So when it comes down, for example, to

let's say something, something concrete,

556

:

like understanding these signals that we

think might be from gravitational waves,

557

:

then I mean, a lot of the problems boil

down to, you know, making sure this is a

558

:

signal and not a background or a noise

source.

559

:

So

560

:

That means, of course, building

experiments that are extremely precise

561

:

measurement devices.

562

:

It also means a lot of modeling of the

various components that go in, and kind of

563

:

both on from the theoretical side and also

from the experimental side.

564

:

And then when you get the data, again, to

cross-check, is this really the type of

565

:

signal that we have kind of...

566

:

Do we have a way, a robust way to

distinguish what we call a signal from

567

:

something that we call a background?

568

:

Take it into account that we might not

have thought of every possible background,

569

:

right?

570

:

So do we kind of really have a telltale

signal of what we think the signal would

571

:

look like, right?

572

:

And typically all these analysis are done

as blind analysis, right?

573

:

So you think about what signal you need to

see in order to be convinced that this is

574

:

what you're looking for before you open

the box and look at your data.

575

:

So that's one challenge.

576

:

more kind of on the data analysis or

experimental side.

577

:

The other challenge may be more on the

theory side.

578

:

So when you're kind of building models,

which extend to standard model of particle

579

:

physics, there's many, many options, and

you need some sort of guiding principle.

580

:

And I mean, if you're lucky, you have data

to guide you, you have some sort of

581

:

anomaly, something you feel like, okay,

here's the weak point, right?

582

:

Here's kind of where you need to poke,

where you need to extend.

583

:

Sometimes you have things like simplicity,

right?

584

:

Which you kind of hope is a good

principle, though, of course, you never

585

:

know that that's a good principle.

586

:

Yeah.

587

:

And recently, that's really been a bit of

a challenge, precisely because the

588

:

standout model works as well as it does.

589

:

There's no...

590

:

I mean, sure, we know we need to explain

dark matter, right?

591

:

But there's many, many possible options

how that dark matter could or could not

592

:

tie into the standard model.

593

:

And there's no very obvious way, like,

there's no obvious weak point at the

594

:

standard model.

595

:

It is not precise weak point.

596

:

I mean, there's a global weakness, things

that cannot explain, but it's kind of not

597

:

quite clear where exactly it needs to be

refined or extended.

598

:

And that I think for

599

:

In the past, it was more clear, or people

had pretty clear ideas, right?

600

:

And then there was pretty obvious things

that needed to be checked, right?

601

:

So we needed to find the Higgs particle,

right?

602

:

So the last missing particle of this then

our model.

603

:

And then we also thought, because the, I

mean, the Higgs particle has certain

604

:

properties, which kind of led us to

believe that we thought, okay, once we

605

:

find the Higgs particle, we should also be

finding other particles somehow related to

606

:

this particle that would naturally explain

certain open challenges.

607

:

But the fact that we haven't found them

and that we're just kind of testing with

608

:

higher and higher accuracy, and we're just

kind of getting the prediction of the

609

:

standard model or confirming the

prediction of the standard model without

610

:

finding any small deviations is making it

very hard to kind of decide a bit.

611

:

What's yeah, how, how should the extension

work?

612

:

Right?

613

:

And how should the extension like is, is

the extension in such a way that we can

614

:

actually test it with.

615

:

with the tools that we have, right?

616

:

Or do we need to think differently?

617

:

I mean, either different types of

experiments, but also maybe different

618

:

theoretical concepts, because so far, most

extensions of the standard model kind of

619

:

rely on the same theoretical framework

point of view theory.

620

:

And then they kind of within that

framework, you try different things.

621

:

But the fact that kind of we haven't had a

real breakthrough there.

622

:

maybe indicating, okay, whatever, you

know, it's just at higher energies, which

623

:

we can't reach, what may be indicating the

framework we're thinking in is maybe not

624

:

the best.

625

:

So yeah, there's many, many questions,

many levels of questions that can be

626

:

addressed.

627

:

Yeah, that's really interesting.

628

:

I'm curious, basically, what would you

like to be true?

629

:

something that at some point nature will

tell you, what would you like to see and

630

:

to observe and the kind of consequences it

would have on our understanding of how the

631

:

universe works?

632

:

Well, I would mainly like nature to

produce something that we can, like give

633

:

us something to work with.

634

:

I would like nature to be kind enough to

produce some sort of signal, be it in dark

635

:

matter, be it in gravitational waves, be

it at a collider.

636

:

that actually gives us something which is

accessible with the two worlds, the

637

:

experiments that we have at the moment.

638

:

Because it could simply be that all these

completions of the standard model live at

639

:

an extremely high energy scale, which is

simply inaccessible to any type of

640

:

collider we can build on Earth.

641

:

And that'll make it not impossible, but

very, very much harder to actually unravel

642

:

these questions.

643

:

Yeah, yeah, for sure.

644

:

And that, I mean, so that's one part of

the work you're doing.

645

:

I told that work around gravitational

waves, which are of course related to

646

:

gravity, in case people didn't understand.

647

:

Oh, and by the way, on the podcast, I had

another researcher called Laura Mansfield

648

:

and she's working on gravity waves.

649

:

which are not the same as gravitational

waves.

650

:

That's quite confusing, but yeah, that's

also actually very interesting field of

651

:

research, basically gravity waves and the

relationship with climate.

652

:

That's all here on Earth.

653

:

But that's also related to gravitational

waves in a way, in the sense that it's big

654

:

objects basically on Earth.

655

:

Like the Everest or the Mont Blanc or all

these big

656

:

massive mountains which actually distort a

bit the gravitational field around them

657

:

and that has impact on the climate.

658

:

How do you model that?

659

:

Basian modeling gets here because that's

really useful because you don't have a lot

660

:

of sample size.

661

:

I recommend listening to Episode 64.

662

:

I put that in the show notes.

663

:

Yeah, I was fascinated by the fact that

gravity, you can study it here on Earth,

664

:

but also it has incredible effects in the

universe and at masses that we cannot even

665

:

imagine, right, with the collisions of

black holes and collisions of neuron

666

:

stars, so that's really something I find

fascinating.

667

:

And actually, can you make the distinction

between a neuron star and a black hole

668

:

listeners and yeah, so that they

understand a bit the difference between

669

:

both.

670

:

Right.

671

:

So a neutron star is made of neutrons,

meaning kind of it's a very, very densely

672

:

packed environment of nuclear matter.

673

:

And a black hole is even more denser,

right?

674

:

So a black hole is really the densest

object that we can imagine.

675

:

where kind of matter has really any type

of matter has really just collapsed into

676

:

this object, and you don't care any much

anymore kind of what it was initially made

677

:

out of, right?

678

:

If we just has one property.

679

:

Of course, it can also spin, but

basically, it only has one property, which

680

:

has which is its mass, right?

681

:

And then it may also have spin if it's if

it's rotating.

682

:

But it doesn't it doesn't matter anymore

what it was made out of.

683

:

So one, one consequence of that is that if

you have two

684

:

neutron stars merging as they get very

close to each other, their gravitational

685

:

force will slightly distort them.

686

:

So they can be a little bit deformed

because despite that they are very, very

687

:

compact, and very dense, they can still be

kind of slightly deformed as they get very

688

:

close to each other, whereas two black

holes will really stay perfectly spherical

689

:

as they as they approach each other.

690

:

So you can tell the difference between the

two by looking at

691

:

details of the gravitational wave signal

as you approach this merger event.

692

:

Okay.

693

:

I didn't know that black hole stayed

spherical even as they approach each

694

:

other.

695

:

Is that because they are so dense that

they cannot be deformed?

696

:

Yeah, it's basically because they are so

dense.

697

:

And because they, I mean, in some sense,

despite that they are physical objects in

698

:

our universe, in some sense, they kind of

become a rather mathematical object.

699

:

Yeah, like a perfect sphere that you

cannot deform or do anything on.

700

:

It's really weird.

701

:

Yeah.

702

:

And it's crazy that we're actually seeing

them, right?

703

:

I mean, both in these gravitation wave

signals as also then with direct

704

:

observations with optical telescopes.

705

:

That's like this first picture of the

black hole in our galaxy and the

706

:

neighboring galaxy.

707

:

Yeah.

708

:

Yeah.

709

:

And so your work on gravity, I'm curious

to understand it because here, obviously

710

:

when we talk about gravity, gravity is so

weak that you have to have so massive

711

:

objects to really see its effects and also

it needs a lot of time.

712

:

So obviously here we're dealing with the

largest scales of the universe.

713

:

But you also work on particle physics, as

you were saying, and you work at CERN,

714

:

where particle physics is one of the

biggest fields.

715

:

So I'm curious, how does that study of

gravity intersect with the study of

716

:

particle physics, especially when we

consider the phenomena you work on, so

717

:

especially black holes and or the early

universe?

718

:

Right.

719

:

Well, I mean, anybody, you know, who's, I

don't know, fallen down the stairs, right,

720

:

will not say gravity is a weak force.

721

:

But indeed, right on Earth, right, when we

compare the force of gravity to the other

722

:

forces that we have, so the forces that

bind atoms together, things like that,

723

:

gravity is extremely weak.

724

:

So when we perform any particle physics

experiment on Earth, we just completely

725

:

neglect gravity, and we're not introducing

any error in our estimations.

726

:

Now, gravity can become important, as you

say, either if you have some very massive

727

:

objects like black holes, or if you have

very far distances, because here on Earth,

728

:

kind of, okay, we have so much matter

interacting so strongly that we don't care

729

:

about gravity.

730

:

But the universe as a whole is actually

pretty empty.

731

:

So in most of the universe, there's just

nothing.

732

:

What leading order, there's nothing.

733

:

And that means that on those scales,

because there's no matter which

734

:

has any interactions that are stronger on

those large scales, it's really gravity

735

:

that is describing the dynamics of the

universe.

736

:

And so if we want to understand both kind

of the dynamics of the universe today, but

737

:

also extrapolating back in past, if we

want to understand the evolution of the

738

:

universe, the birth of the universe, then

we need to understand gravity.

739

:

And one of the big puzzles, for example,

is

740

:

that at the moment observations tell us

that we are in a phase of the universe

741

:

where the universe is not only expanding,

but expanding in an accelerated way.

742

:

And that's pretty weird because normally

you think if you just have a bunch of

743

:

matter, right, a bunch of galaxies, you

think, well, they're going to have

744

:

gravitational interactions between each

other.

745

:

So even if you somehow gave them some

initial velocity, you would think, okay,

746

:

well, they're going to kind of slow down.

747

:

and eventually crunch back together again,

because on those large scales, it's only

748

:

gravity that is important.

749

:

So on those large scales, you think you

can you can either have things collapsing,

750

:

or you can have kind of things, at least

if they're expanding, they should be

751

:

slowing down.

752

:

What we observe is the opposite, right?

753

:

What we observe is really, things are

deferred, things are away from us, the

754

:

faster they are moving away.

755

:

So we're in a universe which is expanding

faster and faster.

756

:

And that is also gravity driving that.

757

:

It's just not the usual form of gravity

that we know on Earth, that gravity is

758

:

attractive.

759

:

But in some sense, you can call it a

repulsive force of gravity, or it's a part

760

:

of gravity that acts as a pressure that

drives the universe apart.

761

:

And that is what we call in dark energy.

762

:

So again, the term dark just implies we

don't really understand and we can't see

763

:

it.

764

:

And energy basically comes from

observations that it has this effect of

765

:

driving the energy of driving the universe

apart.

766

:

So it acts as a type of energy in the

expansion history of our universe and

767

:

concretely today.

768

:

But we don't really so we can model it,

but we can't we don't really fundamentally

769

:

understand what it is.

770

:

So understanding that and understanding

kind of.

771

:

how the universe evolved, not only today,

but in the past.

772

:

That then immediately ties back into

particle physics, because going back in

773

:

time in an expanding universe means you go

to a smaller universe where everything was

774

:

much more dense, much more hot.

775

:

You end up in this primordial soup of

particles.

776

:

So you're looking at particles at high

temperatures, particles when they're

777

:

really kind of not bound in atoms and

molecules, but when they exist really in

778

:

their fundamental

779

:

basically a lab to study particle physics.

780

:

So that's how the connection works between

these very large scales of the universe

781

:

and then the very smallest particles that

we study in that way.

782

:

I see.

783

:

Yeah, it's because then it's because

you're going back to the early universe

784

:

where basically the structure that we have

today of the universe didn't apply because

785

:

it didn't exist yet.

786

:

Correct.

787

:

Correct.

788

:

We go back to when everything was really

kind of just this hot primordial soup of

789

:

fundamental particles.

790

:

We tried to understand kind of how

different properties of the soup, meaning

791

:

different possible extensions of the

standard model, would kind of leave traces

792

:

in the evolution of the universe.

793

:

So would leave traces in kind of

astrophysical and cosmological

794

:

observations that we can make today.

795

:

I see.

796

:

And...

797

:

these days, what's a specific experiment

or project that you're involved in, in

798

:

this film, and what would be the main

question that this project is trying to

799

:

answer?

800

:

Right.

801

:

So a big, big project I'm involved in,

right?

802

:

So this is a, you know, many hundreds,

thousands of people working together is

803

:

the LISA project.

804

:

So that's a future space-based

gravitational wave observatory.

805

:

It's going to be an ESA mission.

806

:

The idea is to have three satellites

circling around the sun on an orbit

807

:

similar to the Earth.

808

:

So following Earth.

809

:

on an orbit around the sun.

810

:

The satellites will be two and a half

million kilometers apart.

811

:

They will exchange laser links.

812

:

So they will be shooting, there will be

lasers going between all combinations of

813

:

the satellites.

814

:

And using these lasers, the idea is to

measure very precisely distance between

815

:

these satellites as they orbit the sun.

816

:

And the idea is that if a gravitational

wave comes, since it's a

817

:

little ripple in space-time, it will

change very slightly the distance between

818

:

the satellites.

819

:

And so by kind of looking for this,

looking for these little variations in the

820

:

distance between the satellites, the goal

is to look for gravitational waves.

821

:

And being in space has the big advantage

that a lot of the noise that you have to

822

:

deal with on Earth is not there.

823

:

So the idea is that you can

824

:

much better sensitivities than you could

on Earth.

825

:

Yeah, that makes sense.

826

:

Also, although I'm guessing the sun can be

noisier at times.

827

:

Right, but it's all a question of

frequency, right?

828

:

So you need to kind of find a frequency

band which is clean.

829

:

But yeah, I mean, there's obviously huge

technological challenges in implementing a

830

:

mission like this and many things that can

go wrong.

831

:

This is why you need a lot of people with

a lot of different expertise coming

832

:

together and also a lot of money to build

an instrument like that.

833

:

Yeah.

834

:

I mean, just the engineering part of it is

you have to launch three satellites.

835

:

First, that's already hard.

836

:

And then you have to put them in orbit

around the sun and that they still can

837

:

communicate with each other.

838

:

It's just, and they are extremely far

apart from each other.

839

:

So just that part is...

840

:

absolutely incredible that we can do that.

841

:

Knock, knock, right?

842

:

I mean, we hope we can do it.

843

:

Yeah, I mean, that's just incredibly

fascinating.

844

:

And so what's the ETA on this mission?

845

:

When will the satellites go up

theoretically?

846

:

Right.

847

:

So the hope is to launch in the early

:

848

:

but it's really not.

849

:

Because, yeah, I mean, it takes a while to

build a satellite.

850

:

And also to develop all the kind of the

data analysis pipelines that you need.

851

:

Make sure you have all the sensors on

board that you might need to perform

852

:

whatever type of cross checks.

853

:

Yeah, make sure you didn't put anything on

board, which generates a bunch of noise.

854

:

Because once it's up there, it's up there,

right?

855

:

You can't.

856

:

Yeah.

857

:

Yeah, I mean, it's not in the orbit,

right?

858

:

Exactly.

859

:

You cannot find it, send anybody to repair

it, right?

860

:

So once it's up there, it's up there.

861

:

So you really have to think of every

possible complication beforehand.

862

:

Yeah, which is quite daunting.

863

:

I have to do that for my own statistical

model, you know, where I probe them and

864

:

I'm like, okay, where can the model fail?

865

:

What could be the potential issues?

866

:

It's already...

867

:

stressing me out, but then if you have to

do that for something you cannot go back

868

:

to, that's just incredibly daunting.

869

:

If you think a code release is stressful,

then imagine this.

870

:

Oh, yeah.

871

:

Oh my God.

872

:

But so fascinating.

873

:

Personally, what's your part in this

project, for instance, in the Lisa

874

:

project?

875

:

Right.

876

:

I'm in charge of coordinating research on

what we call

877

:

the stochastic backgrounds.

878

:

So the signals we've talked about so far,

and predicted the ones we see by LIGO, are

879

:

what we call transient signals, meaning

most of the time the detector actually

880

:

sees nothing, just noise.

881

:

And then from time to time, you have a

rather relatively strong signal.

882

:

You see it, then it's gone.

883

:

So if that's your data analysis challenge,

then you can calibrate your detector in

884

:

the signal-free moments.

885

:

You can learn all about your properties of

the noise and you can have a good noise

886

:

model.

887

:

And then when you get a signal, you can

kind of do a pretty good signal to noise

888

:

discrimination.

889

:

Now with Lisa, the situation is going to

be very different because we're going to

890

:

have, because it's such a sensitive

instrument, we're going to have lots and

891

:

lots of stuff going on all the time.

892

:

So we're basically not going to have

signal free time.

893

:

So we're kind of.

894

:

dealing with kind of measuring all these

different signals and the noise at the

895

:

same time.

896

:

And at the same time, the idea is that we

might have stochastic backgrounds.

897

:

So stochastic backgrounds could, they're

not transient signals, but there's kind of

898

:

more like a white noise, which is there at

all times.

899

:

They could be coming from unresolved

astrophysical sources, so unresolved black

900

:

or black or merges that are kind of out of

the range of our detector.

901

:

So we can't individually detect them, but

they just kind of contribute to some

902

:

confusion noise.

903

:

Or they could be these signals from the

very early universe, which is, of course,

904

:

the ones that I'm actually after.

905

:

But so you have to kind of dig them out

between all these loud transient signals,

906

:

between these possible astrophysical noise

like signals, which look very, very

907

:

similar to the kind of cosmological noise

like signal that you will be looking for.

908

:

And of course, the words are very, very

similar to instrument noise that you might

909

:

have mismodeled or misunderstood.

910

:

So.

911

:

And what I'm working on is okay, a on on,

okay, understanding the possible models

912

:

for these for these different components,

in particular for the cosmological

913

:

sources, but also trying to understand how

could we if we you know, actually get some

914

:

actual data, how can we actually

disentangle all of these components?

915

:

And how can we really kind of make the

most of the of the mission, extract as

916

:

much information as possible?

917

:

which with all these kind of overlapping

signals and challenges.

918

:

Yeah, yeah.

919

:

And I'm guessing that having to do that,

not in a few months is something you

920

:

appreciate.

921

:

Yes.

922

:

Yes, yes, yes.

923

:

Yeah, so there's many challenges out

there.

924

:

Obviously, many people working on it.

925

:

And I mean, luckily, as you say, luckily,

we don't have to solve this in a couple of

926

:

months, right?

927

:

Because we're basically also counting on

things like computing power, and so on,

928

:

increasing new methods becoming available.

929

:

But, but yeah, so it's, but still, I mean,

the development has to happen now.

930

:

Because if we kind of figure, okay, we

need a certain type of

931

:

sensor or some certain type of output data

that would help us to discriminate these

932

:

different signals.

933

:

We can't come along with that when the

mission is already built or even worse,

934

:

already launched.

935

:

So you can't wait till you see the data to

decide how you're going to do the

936

:

analysis.

937

:

You at least have to have a very good idea

of how you're going to do the analysis

938

:

before you see the data.

939

:

And then maybe you can refine once you see

the data.

940

:

Yeah, definitely.

941

:

Actually, this kind of work that you do in

theoretical physics or that kind of

942

:

project you just described, it really

involves the development of models, of

943

:

hypotheses, and I'm curious if you have

some favorite hypotheses or models or the

944

:

most intriguing theoretical ideas.

945

:

that you've encountered in your field and

that you'd like to see tested.

946

:

And if we could actually test them right

now with our current technology.

947

:

Good question.

948

:

I must say, I don't have a particularly

favorite model.

949

:

I don't feel, I don't know, protective

ownership of any particular idea.

950

:

I'm more the type of person who I start

working on something because I find it

951

:

interesting.

952

:

And then once I've understood it to a

certain degree, I move on to the next

953

:

topic.

954

:

But I think there are a couple of kind of

big overarching...

955

:

questions, right?

956

:

So kind of, yeah, understanding, getting

some experimental input on what on what

957

:

dark matter is, would really help a lot on

the on the theory development side.

958

:

As I mentioned, when we also have issues

understanding the Higgs particle,

959

:

understanding in particular mass of the

Higgs particle, which is potentially

960

:

indicating there's something we don't

understand properly about quantum field

961

:

theory about

962

:

that I find is incredibly exciting,

because it would really mean kind of,

963

:

okay, not an add on, you know, not a small

extension of our existing model, but

964

:

really, completely revolution and how we

think about things.

965

:

Yeah, of course, it also makes it much

more difficult, right?

966

:

Because you don't even have the framework.

967

:

Maybe we don't even have the mathematical

framework to think about this.

968

:

It's a huge step to take.

969

:

So I would, I mean, that's what would be a

big step, right?

970

:

So I'm not sure if and how that's going to

happen.

971

:

If it's even necessary, right?

972

:

Maybe the current framework is totally

fine, but that would definitely be a

973

:

development that on just on the pure

theory side, that would be very exciting

974

:

to see happening.

975

:

Yeah.

976

:

Yeah, for sure.

977

:

Definitely.

978

:

I kind of, I'm also really curious about

that.

979

:

Actually, is there one big question that

you would like to see answered before you

980

:

die?

981

:

Your one big question that you'd really

like the answer to.

982

:

I think I really would like to know the

answer to Dark Matter.

983

:

Just because that-

984

:

It's well, there's this we have many, we

have many very reasonable models, which

985

:

can be tested and which are being tested.

986

:

So we could still be unlucky and nature

could choose not one of these nice and

987

:

reasonable models, right, but something

completely different.

988

:

But that that's a field where there are

some very good suggestions and they can be

989

:

tested.

990

:

Now, unfortunately, there was one

excellent suggestion, right, which was

991

:

supersymmetry and the dark matter particle

that comes with supersymmetry would have

992

:

solved, was mathematically beautiful,

would have solved a ton of questions, was

993

:

in many ways the perfect theory, right?

994

:

Unfortunately, we didn't find it.

995

:

So it could still be out there, but kind

of not as a solution to all of the

996

:

problems that we hoped it would solve.

997

:

Because if that were the case, we should

already have seen it.

998

:

Yeah, so something kind of being the ideal

theory from our point of view, doesn't

999

:

mean nature actually cares, right?

:

01:04:54,343 --> 01:04:55,224

Yeah, for sure.

:

01:04:55,224 --> 01:04:56,525

And does it that way.

:

01:04:58,126 --> 01:05:04,090

But yeah, so Dark Matter, I think it

really has the potential that we could

:

01:05:04,090 --> 01:05:05,171

actually find it.

:

01:05:05,171 --> 01:05:10,795

And if we find it, that could really be a

starting point of a whole new exploration

:

01:05:10,795 --> 01:05:13,216

of questions.

:

01:05:13,216 --> 01:05:15,057

Yeah, definitely.

:

01:05:15,870 --> 01:05:22,575

And that's interesting that you mentioned

dark matter too, because Kevin Clive, I

:

01:05:22,575 --> 01:05:26,498

asked him the same question and he

answered dark matter too.

:

01:05:26,498 --> 01:05:30,662

So that's interesting to see that it's

really something that's picking up in the

:

01:05:30,662 --> 01:05:38,628

physics space these days where it seems

like we're less, let's say we're more

:

01:05:38,628 --> 01:05:45,398

hopeful that we can actually start making

sense of it and probing

:

01:05:45,398 --> 01:05:50,339

the universe in a way that will give us

some answers, at least to this mystery.

:

01:05:50,799 --> 01:05:54,500

Whereas dark energy, from what I

understand, we understand way less about

:

01:05:54,500 --> 01:05:57,921

dark energy than we understand about dark

matter for now, right?

:

01:05:57,921 --> 01:05:59,221

Yeah.

:

01:05:59,221 --> 01:05:59,821

That's correct.

:

01:05:59,821 --> 01:06:06,283

And also there we have much less, I mean,

we see what it does on large scales,

:

01:06:06,283 --> 01:06:06,563

right?

:

01:06:06,563 --> 01:06:14,185

But we have also much less of an idea how

to make progress.

:

01:06:14,518 --> 01:06:19,641

Both on the theory side, there's kind of

not these kind of clear cut models that

:

01:06:19,641 --> 01:06:23,783

kind of say, okay, here's a good theory of

why it is how it is, and here's how you go

:

01:06:23,783 --> 01:06:25,764

test it, right?

:

01:06:25,764 --> 01:06:26,905

For Dark Energy, we have neither.

:

01:06:26,905 --> 01:06:31,307

Neither a clear cut theory that kind of

says, okay, here's a good explanation, nor

:

01:06:31,307 --> 01:06:33,769

any way of probing them really.

:

01:06:34,429 --> 01:06:37,551

So it's a much, it's much more in the

blur.

:

01:06:37,551 --> 01:06:38,151

Yeah.

:

01:06:40,753 --> 01:06:41,813

So hopefully.

:

01:06:41,886 --> 01:06:47,211

In 10 days, you'll come back to the show

and we'll talk about Dark Energy and the

:

01:06:47,211 --> 01:06:51,294

latest progresses.

:

01:06:51,354 --> 01:06:56,199

Valerie, I think I have so many more

questions, but you've been already very

:

01:06:56,199 --> 01:06:57,820

generous with your time.

:

01:06:58,701 --> 01:07:03,085

Before closing up, is there any topic I

didn't ask you about and that you'd like

:

01:07:03,085 --> 01:07:03,905

to mention?

:

01:07:05,786 --> 01:07:10,170

I think we covered a lot, but nothing

particular comes to my mind.

:

01:07:10,170 --> 01:07:11,552

Okay.

:

01:07:11,552 --> 01:07:18,819

Well, then I think we can call it a show,

but as usual, before I think you go, I'm

:

01:07:18,819 --> 01:07:23,163

going to ask you the last two questions I

ask every guest at the end of the show.

:

01:07:23,484 --> 01:07:28,729

First one, if you had unlimited time and

resources, which problem would you try to

:

01:07:28,729 --> 01:07:29,569

solve?

:

01:07:32,883 --> 01:07:38,732

Yeah, that's as I said, that's actually a

really tricky question because we are in

:

01:07:38,732 --> 01:07:45,221

this in this situation that I find it very

hard to pinpoint.

:

01:07:47,734 --> 01:07:49,495

where is the weak point of the standard

model?

:

01:07:49,495 --> 01:07:51,376

Where should we poke it?

:

01:07:51,376 --> 01:07:51,596

Right?

:

01:07:51,596 --> 01:08:01,121

So from the pure theory side, without any

experimental input, I feel like if I had

:

01:08:01,121 --> 01:08:06,184

unlimited time and resources, I wouldn't

engage on a single project right now.

:

01:08:08,365 --> 01:08:15,189

But I would basically just try and, you

know, gather as broad as possible

:

01:08:15,189 --> 01:08:16,609

understanding of

:

01:08:17,042 --> 01:08:23,643

as many concepts as possible and hope that

we will eventually get some sort of data,

:

01:08:23,643 --> 01:08:26,764

which points us in the direction we need

to explore.

:

01:08:26,764 --> 01:08:31,105

I don't at the moment really have a clear

cut avenue where I say this is where I

:

01:08:31,105 --> 01:08:32,366

would put all my money.

:

01:08:35,487 --> 01:08:37,107

Yeah.

:

01:08:37,107 --> 01:08:43,469

So wise answer where you don't put your

eggs in the same basket.

:

01:08:43,469 --> 01:08:45,610

And second question, if you could have

dinner.

:

01:08:45,610 --> 01:08:51,975

with any great scientific mind, dead,

alive or fictional, who would it be?

:

01:08:51,975 --> 01:08:54,677

Yeah, I think, well, we'd go for somebody

dead, right?

:

01:08:54,677 --> 01:08:58,360

Just because that's a chance you don't get

on a regular conference dinner.

:

01:08:59,761 --> 01:09:05,586

So I'd be really curious to talk with some

of the people involved in the discovery of

:

01:09:05,586 --> 01:09:06,927

quantum mechanics.

:

01:09:07,127 --> 01:09:09,689

So say Heisenberg or somebody like that.

:

01:09:10,290 --> 01:09:14,873

Because I feel like they were kind of...

:

01:09:15,558 --> 01:09:21,723

at the core of the field, when the field

was also in a situation where it was kind

:

01:09:21,723 --> 01:09:26,567

of not so clear cut, at that time, not

even clear cut that it was a need to kind

:

01:09:26,567 --> 01:09:30,851

of extend the current understanding

because classical physics was well

:

01:09:30,851 --> 01:09:31,631

understood, right?

:

01:09:31,631 --> 01:09:35,875

And nearly all phenomena were very well

understood.

:

01:09:35,875 --> 01:09:39,238

And people were thinking, okay, you know,

physics, it's done, you know, we

:

01:09:39,238 --> 01:09:40,418

understand nature.

:

01:09:41,439 --> 01:09:44,014

And it was just kind of very small.

:

01:09:44,014 --> 01:09:47,035

tweaks here and there, right, that kind of

were a bit confusing.

:

01:09:48,316 --> 01:09:52,558

So one could have easily believed

everything is done and understood, go

:

01:09:52,558 --> 01:09:53,919

study something else.

:

01:09:54,720 --> 01:09:58,482

But they kind of opened the door to the

world of quantum physics.

:

01:09:59,202 --> 01:10:04,345

And with that then came quantum field

theory, with that came kind of elementary

:

01:10:04,345 --> 01:10:08,908

particle physics, with that came kind of

all the questions that we have today.

:

01:10:09,488 --> 01:10:12,149

So actually, from today's point of view,

:

01:10:12,522 --> 01:10:14,803

we would say, well, they understood very

little, right?

:

01:10:14,803 --> 01:10:20,006

It was a whole bunch of new physics that

was kind of not known to them, but they

:

01:10:20,006 --> 01:10:22,807

didn't even know that it was not known to

them, because there was kind of no glaring

:

01:10:22,807 --> 01:10:23,888

open question.

:

01:10:24,828 --> 01:10:30,512

So I'd really be curious to know how they

perceived that situation and how they got

:

01:10:30,512 --> 01:10:34,574

to the point of opening the door to the

quantum world and taking up that

:

01:10:34,574 --> 01:10:34,954

challenge.

:

01:10:34,954 --> 01:10:37,275

Yeah, yeah, yeah.

:

01:10:37,275 --> 01:10:40,997

Yeah, definitely sounds like a very fine

dinner.

:

01:10:41,602 --> 01:10:44,502

Please invite me.

:

01:10:44,502 --> 01:10:46,003

So, well, awesome.

:

01:10:46,003 --> 01:10:48,123

Thanks a lot, Varyry.

:

01:10:48,123 --> 01:10:50,564

That was absolutely fascinating.

:

01:10:51,084 --> 01:10:54,685

We didn't talk a lot about stats, but I

love doing these episodes from time to

:

01:10:54,685 --> 01:11:00,527

time, you know, where we de-zoom a bit

from stats and just talk about fascinating

:

01:11:00,527 --> 01:11:02,148

science in general.

:

01:11:02,808 --> 01:11:08,529

I think it's very interesting and also

quite important to put more rigorous

:

01:11:09,338 --> 01:11:12,741

pedagogical scientific content out there

in the world.

:

01:11:12,781 --> 01:11:14,402

We've seen that in the recent years.

:

01:11:14,402 --> 01:11:19,046

So thanks a lot for doing this for us,

Valérie.

:

01:11:20,027 --> 01:11:23,690

I will put a link to your website in the

show notes for those who want to dig

:

01:11:23,690 --> 01:11:24,570

deeper.

:

01:11:24,631 --> 01:11:30,676

Also feel free to add any link to cool

papers or experiments or videos that you

:

01:11:30,676 --> 01:11:32,998

think listeners will appreciate.

:

01:11:33,298 --> 01:11:37,621

And thank you again, Valérie, for taking

the time and being on this show.

:

01:11:38,582 --> 01:11:39,462

Thank you.

:

01:11:39,542 --> 01:11:44,684

And rest assured that stats is still at

the basis of all this, despite that we

:

01:11:44,684 --> 01:11:48,025

took a more high-level approach in this

discussion.

:

01:11:48,345 --> 01:11:50,486

Yeah, for sure.

:

01:11:50,486 --> 01:11:56,788

Well, thanks a lot, Valerie, and see you

soon on the show.

Chapters

Video

More from YouTube