PageSourceSearch

https://rubenohana.github.io/

html rubenohana.github.io collected 2026-10-03 09:52:16 UTC 27,804 bytes, 452 lines download raw bytes

1<!DOCTYPE html>
2<!--
3  flatironinstitute.org
4  -->
5<html lang="en">
6  <head>
7    <title>Ruben Ohana, Ph.D. | Generative AI Research Scientist</title>
8    <meta charset="utf-8" />
9    <meta
10      name="viewport"
11      content="width=device-width, initial-scale=1"
12    />
13    <meta
14      name="description"
15      content="Personal website of Ruben Ohana, Ph.D., Senior Research Scientist in Generative AI at NVIDIA."
16    />
17    <link rel="canonical" href="https://rubenohana.github.io/" />
18    <meta property="og:title" content="Ruben Ohana, Ph.D." />
19    <meta
20      property="og:description"
21      content="Senior Research Scientist in Generative AI at NVIDIA."
22    />
23    <meta property="og:type" content="website" />
24    <meta property="og:url" content="https://rubenohana.github.io/" />
25    <!-- TODO: If you wish to use your center color scheme, please replace main.css with the center acronymn ex:
26      cca.css, ccb.css, etc
27    -->
28    <link rel="stylesheet" href="assets/css/ccm.css" />
29    <link rel="icon" type="image/svg+xml" href="images/favicon.svg" />
30  </head>
31  <body class="is-preload">
32    <!-- Header -->
33    <header id="header">
34      <div class="inner">
35        <!-- TODO: Add a personal photo in /images folder or switch src link to avatar.jpg for a gray circle -->
36        <a href="#" class="image avatar"
37          ><img src="images/Headshot.jpeg" alt="Portrait of Ruben Ohana"
38        /></a>
39      </div>
40      <nav id="nav">
41        <ul>
42          <li>
43            <a
44              href="#about"
45              style="-webkit-tap-highlight-color: rgba(0, 0, 0, 0)"
46              >About</a
47            >
48          </li>
49          <li>
50            <a
51              href="#research"
52              style="-webkit-tap-highlight-color: rgba(0, 0, 0, 0)"
53              >Research</a
54            >
55          </li>
56          <li>
57            <a
58              href="#publications"
59              style="-webkit-tap-highlight-color: rgba(0, 0, 0, 0)"
60              >Publications</a
61            >
62          </li>
63          <li>
64            <a
65              href="#talks"
66              style="-webkit-tap-highlight-color: rgba(0, 0, 0, 0)"
67              >Talks</a
68            >
69          </li>
70          <!-- <li>
71            <a
72              href="#software"
73              style="-webkit-tap-highlight-color: rgba(0, 0, 0, 0)"
74              >Software</a
75            >
76          </li> -->
77          <li>
78            <a
79              href="#contact"
80              style="-webkit-tap-highlight-color: rgba(0, 0, 0, 0)"
81              >Contact</a
82            >
83          </li>
84          <!-- <ul class="icons"> -->
85            <!-- TODO: Add your social media handles here -->
86            <!-- Other icons are available here: https://fontawesome.com/v4.7/icons/ just change the fa-tag -->
87            <!-- <li>
88              <a href="#" class="icon brands fa-twitter"
89                ><span class="label">Twitter</span></a
90              > 
91              <a href="#" class="icon brands fa-google-scholar"
92                ><span class="label">Instagram</span></a
93              >
94            </li> -->
95          <!-- </ul> -->
96
97        </ul>
98        
99      </nav>
100    </header>
101    <!-- Main -->
102    <div id="main">
103      <!-- About -->
104      <section id="about">
105        <header class="major">
106          <!-- TODO: Add your name -->
107          <h1>Ruben Ohana, Ph.D.</h1>
108        </header>
109        <!-- TODO: Add your own bio -->
110        <p>
111          I am a Senior Research Scientist in Generative AI at <a href="https://www.nvidia.com/en-us" rel="noreferrer noopener" target="_blank">NVIDIA</a> in Santa Clara, CA. Previously, I was a Research Fellow at the <a href="https://www.simonsfoundation.org/flatiron/center-for-computational-mathematics/" rel="noreferrer noopener" target="_blank">Center for Computational Mathematics</a> of the <a href="https://www.simonsfoundation.org/flatiron/" rel="noreferrer noopener" target="_blank">Flatiron Institute</a> in New York, and a member of the <a href="https://www.polymathic-ai.org" rel="noreferrer noopener" target="_blank">Polymathic AI</a> collaboration. I received my Ph.D. in Computer Science from the Physics Lab of <a href="https://www.ens.psl.eu">Ecole Normale Supérieure</a>
111 in 2022, and was at the same time a Research Scientist for the startup <a href="https://lighton.ai/">LightOn</a>, where I worked on developing new machine learning algorithms and applications (adversarial robustness, differential privacy, time-series prediction...) for Optical Processing Units. During my PhD, I interned at the <a href="https://ailab.criteo.com/">Criteo AI Lab</a>.
112        </p>
113        <p>
114          My academic research focused on building foundation models for scientific data (mostly numerical simulations of partial differential equations and astrophysics data), as well as dataset building and on training large deep learning models, as part of the Polymathic AI collaboration.
115        </p>
116        <p>
117          Before my Ph.D., I graduated with an engineering degree from <a href="https://www.espci.psl.eu/en/">ESPCI Paris</a>, an MSc in Condensed Matter from <a href="https://www.ens.psl.eu">Ecole Normale Supérieure</a>, and an MSc in Statistics from <a href="https://www.sorbonne-universite.fr/en/university">Sorbonne University</a>. During these undergraduate studies, I interned at <a href="https://www.rd.ntt/e/brl/">NTT Basic Research Laboratory</a> in Japan, the <a href="https://www.ligo.caltech.edu/">LIGO</a> at <a href="https://www.mit.edu/">M.I.T.</a> and the <a href="https://www.lip6.fr/recherche/team.php?LANG=en&amp;acronyme=QI">Quantum Information Group</a> at the <a href="https://www.lip6.fr/?LANG=en">LIP6</a> at Sorbonne University.
118        </p>
119        <p>
120          You can find more information about me here: <a href="pdfs/CV.pdf" rel="noreferrer noopener" target="_blank">CV</a> -
121          <a href="https://scholar.google.fr/citations?user=F9qNg2wAAAAJ&hl=en" rel="noreferrer noopener" target="_blank">Google Scholar</a> -
122          <a href="https://twitter.com/oharub" rel="noreferrer noopener" target="_blank">
123            <i class="fab fa-twitter"></i>
124          </a> - 
125          <a href="https://www.linkedin.com/in/rubenohana/" rel="noreferrer noopener" target="_blank">
126            <i class="fab fa-linkedin"></i>
127          </a> -
128        <a href="https://github.com/rubenohana" rel="noreferrer noopener" target="_blank">
129            <i class="fab fa-github"></i>
130          </a>
131        </p>
132      </section>
133      <!-- Research -->
134      <section id="research">
135        <h2>Spotlight Research</h2>
136        <!-- TODO: Replace with project info and choose image type -->
137        <div class="row">
138          <article class="col-6 col-12-xsmall work-item">
139            <h3>The Well: a large-scale collection of physics numerical simulations for machine learning</h3>
140            <p>
141              We released the most extensive collection of physics numerical simulations datasets, which comprises numerical simulations of diverse physics phenomena ranging from biology to astrophysics. We hope it will advance both the developement of ML surrogate models, accelerate scientific discovery, and allow the training of foundation models. The code can be found <a href="https://github.com/PolymathicAI/the_well" rel="noreferrer noopener" target="_blank">here</a> and you can read more about it in our <a href="https://arxiv.org/abs/2412.00568" rel="noreferrer noopener" target="_blank"> NeurIPS 2024</a> paper.
142            </p>
143          </article>
144          <article class="col-6 col-12-xsmall work-item">
145            <img class="image fit" src="images/biggif.gif" alt="Visualization for The Well project" />
146          </article>
147           <article class="col-6 col-12-xsmall work-item">
148            <h3>Blind denoising using Gibbs-Diffusion.</h3>
149            <p>
150              We developed a very cool blind denoising algorithm for natural images, with an application in astrophysics. It can at the same time do parameter inference with uncertainties. We believe it has the potential to be applied in many different scientific fields. The project was accepted at <a href="https://arxiv.org/abs/2402.19455" rel="noreferrer noopener" target="_blank">ICML 2024</a> and the code can be found
151              <a
152                rel="noreferrer noopener"
153                target="_blank"
154                href="https://github.com/rubenohana/Gibbs-Diffusion"
155                >here.</a
156              >
157            </p>
158          </article> 
159          <article class="col-6 col-12-xsmall work-item">
160            <img
161              class="image fit"
162              src="images/denoising_effect2.gif"
163              alt="Example of Gibbs-Diffusion image denoising"
164            />
165          </article>
166
167        </div>
168      </section>
169      <!-- Papers -->
170      <section id="publications">
171        <h2>Publications</h2>
172        <div class="row">
173          <div class="col-12 col-12-small">
174            <!-- TODO: Add your citations  -->
175            <ol>
176
177              <li><i>Streamlined Optical Training of Large-Scale Modern Deep Learning Architectures with Direct Feedback Alignment.</i> Z. Wang, K. Müller, M. Filipovich, J. Launay, <b>R. Ohana</b>, G. Pariente, S. Mokaadi, C. Brossollet, F. Moreau, A. Cappelli, I. Poli, I. Carron, L. Daudet, F. Krzakala, S. Gigan. <a href="https://doi.org/10.1073/pnas.2532022123" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">PNAS 2026</a></li>
178
179              <li><i>On the Value of Tokeniser Pretraining in Physics Foundation Models.</i> H. Sotoudeh, P. Mukhopadhyay, <b>R. Ohana</b>, M. McCabe, N. D. Lawrence, S. Ho, M. Cranmer. <a href="https://arxiv.org/abs/2603.05598" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">ICLR 2026, AI &amp; PDE Workshop</a></li>
180
181              <li><i>Overtone: Cyclic Patch Modulation for Clean, Efficient, and Flexible Physics Emulators.</i> P. Mukhopadhyay, M. McCabe, <b>R. Ohana</b>, M. Cranmer. <a href="https://arxiv.org/abs/2507.09264" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">ICLR 2026</a></li>
182
183              <li><i>Walrus: A Cross-Domain Foundation Model for Continuum Dynamics.</i> <b>Polymathic AI.</b> <a href="https://arxiv.org/abs/2511.15684" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">ICML 2026 (Spotlight)</a></li>
184
185              <li><i>AION-1: Omnimodal Foundation Model for Astronomical Sciences.</i> <b>Polymathic AI.</b> <a href="https://arxiv.org/abs/2510.17960" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">NeurIPS 2025</a></li>
186
187              <li><i>TC-LoRA: Temporally Modulated Conditional LoRA for Adaptive Diffusion Control.</i> M. Cho, <b>R. Ohana</b>, C. Jacobsen, A. Jothi, M.-H. Chen, Z. Morley Mao, E. Can. <a href="https://arxiv.org/abs/2510.09561" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">NeurIPS 2025, Space in VLEAI Workshop</a></li>
188
189              <li><i>Lost in Latent Space: An Empirical Study of Latent Diffusion Models for Physics Emulation.</i> F. Rozet, <b>R. Ohana</b>, M. McCabe, G. Louppe, F. Lanusse, S. Ho. <a href="https://arxiv.org/abs/2507.02608" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">NeurIPS 2025</a></li>
190
191              <li><i>
191Predicting Partially Observable Dynamical Systems via Diffusion Models with a Multiscale Inference Scheme.</i> R. Morel, F. P. Ramunno, J. Shen, A. Bietti, K. Cho, M. Cranmer, S. Golkar, O. Gugnin, G. Krawezik, T. Marwah, M. McCabe, L. Meyer, P. Mukhopadhyay, <b>R. Ohana</b>, L. Parker, H. Qu, F. Rozet, K. D. Leka, F. Lanusse, D. Fouhey, S. Ho. <a href="https://arxiv.org/abs/2511.19390" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">NeurIPS 2025</a></li>
192
193              <li>
194                <i>The Well: a Large-Scale Collection of Diverse Physics Simulations for Machine Learning. </i> <b>R. Ohana and M.McCabe, Polymathic AI. </b>
195                <!-- TODO: Add external links -->
196                <a
197                  href="https://arxiv.org/abs/2412.00568"
198                  rel="noreferrer noopener"
199                  target="_blank"
200                  class="button primary icon xsmall solid fa-external-link-alt"
201                  >NeurIPS 2024 D&B track.</a
202                >
203              </li>
204
205              <li>
206                <i>Multiple Physics Pretraining for Spatiotemporal Surrogate Models. </i> <b>Polymathic AI.</b>
207                <!-- TODO: Add external links -->
208                <a
209                  href="https://arxiv.org/abs/2310.02994"
210                  rel="noreferrer noopener"
211                  target="_blank"
212                  class="button primary icon xsmall solid fa-external-link-alt"
213                  >NeurIPS 2024.</a
214                >
215              </li>
216
217              <li>
218                <i>Listening to the Noise: Blind Denoising with Gibbs Diffusion </i> D. Heurtel-Depeiges, C. Margossian, R. Ohana, B. Régaldo.   
219                <!-- TODO: Add external links -->
220                <a
221                  href="https://arxiv.org/abs/2402.19455"
222                  rel="noreferrer noopener"
223                  target="_blank"
224                  class="button primary icon xsmall solid fa-external-link-alt"
225                  >ICML 2024</a
226                >
227              </li>
228
229              <li>  <i>MoMo: Momentum Models for Adaptive Learning Rates.</i> F. Schaipp, <b>R. Ohana</b>, M. Eickenberg, A. Defazio, R. M. Gower. <a href="https://arxiv.org/abs/2305.07583" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">ICML 2024</a>   </li> 
230
231
232              <li>
233                <i>xVal: A Continuous Number Encoding for Large Language Models.  </i> <b>Polymathic AI.</b>
234                <!-- TODO: Add external links -->
235                <a
236                  href="https://arxiv.org/abs/2310.02989"
237                  rel="noreferrer noopener"
238                  target="_blank"
239                  class="button primary icon xsmall solid fa-external-link-alt"
240                  >NeurIPS, AI4Science Workshop 2023.</a
241                >
242              </li>
243              
244              <li>  <i>Removing Dust from CMB Observations with Diffusion Models.</i> D. Heurtel-Depeiges, B. Burkhart, R. Ohana, B. Régaldo-Saint Blancard. <a href="https://arxiv.org/abs/2310.16285" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt"><u>Oral</u> @ NeurIPS, MLPS Workshop 2023.</a>   </li>
245              
246              
247              <li>  <i>AstroCLIP: a cross-modal foundation model for galaxies.</i> <b>Polymathic AI.</b> <a href="https://academic.oup.com/mnras/article/531/4/4990/7697182" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">MNRAS 2024</a>   </li>
248              
249              
250              <li>  <i>Shedding a PAC-Bayesian Light on Adaptive Sliced-Wasserstein Distances.</i> <b>R. Ohana<sup>*</sup></b>, K. Nadjahi<sup>*</sup>, A. Rakotomamonjy, L. Ralaivola. <a href="https://arxiv.org/abs/2206.03230" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">ICML 2023</a>   </li> 
251              
252              <li>  <i>Linear Optical Random Projections Without Holography.</i> <b>R. Ohana</b>
252, D. Hesslow, D. Brunner, S. Gigan, K. Müller. <a href="https://arxiv.org/abs/2305.12988" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">Optics Express</a>   </li> 
253              
254              <li>  <i>Complex-to-Real Sketches for Tensor Products with Applications to the Polynomial Kernel.</i> J. Wacker, <b>R. Ohana</b>, M. Filippone. <a href="https://proceedings.mlr.press/v206/wacker23a/wacker23a.pdf" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">AISTATS 2023</a>   </li>
255              <li>  <i>Adversarial Robustness by Design through Analog Computing and Synthetic Gradients.</i> <b>R. Ohana<sup>*</sup></b>, A. Cappelli<sup>*</sup>, J. Launay, L. Meunier, I. Poli, F. Krzakala. <a href="https://arxiv.org/abs/2101.02115" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">ICASSP 2022</a>   </li> 
256
257              <li>  <i>Photonic Differential Privacy with Direct Feedback Alignment.</i> <b>R. Ohana<sup>*</sup></b>, H.J. Ruiz<sup>*</sup>, J. Launay<sup>*</sup>, A. Cappelli, I. Poli, L. Ralaivola, A. Rakotomamonjy. <a href="https://proceedings.neurips.cc/paper/2021/file/b8c4c8b2271787e2f78b5fe2ce193caa-Paper.pdf" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">NeurIPS 2021</a>   </li>
258
259              <li><i>LightOn Optical Processing Unit: Scaling-up AI and HPC with a Non von Neumann Co-processor.</i> C. Brossollet, A. Cappelli, I. Carron, C. Chaintoutis, A. Chatelain, L. Daudet, S. Gigan, D. He
259sslow, F. Krzakala, J. Launay, S. Mokaadi, F. Moreau, K. Müller, <b>R. Ohana</b>, G. Pariente, I. Poli, E. Tommasone. <a href="https://arxiv.org/abs/2107.11814" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">arXiv 2021</a></li>
260
261              <li>  <i>ROPUST: Improving Robustness through Fine-tuning with Photonic Processors and Synthetic Gradients.</i> A. Cappelli, J. Launay, L. Meunier, <b>R. Ohana</b>, I. Poli. <a href="https://arxiv.org/abs/2108.04217" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">arXiv</a>   </li>
262
263              <li>  <i>Photonic co-processors in HPC: using LightOn OPUs for Randomized Numerical Linear Algebra.</i> D. Hesslow, A. Cappelli, I. Carron, L. Daudet, R. Lafargue, K. Müller, <b>R. Ohana</b>, G. Pariente, I. Poli. <a href="https://arxiv.org/abs/2104.14429" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">Hot Chips 2021</a>   </li>
264
265              <li>  <i>Align, then memorise: the dynamics of learning with feedback alignment.</i> M. Refinetti<sup>*</sup>, S. d'Ascoli<sup>*</sup>, <b>R. Ohana</b>, S. Goldt. <a href="https://arxiv.org/abs/2011.12428" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">ICML 2021</a>   </li>
266
267              <li><i>Binarization for Optical Processing Units via REINFORCE.</i> B. Kozyrskiy, I. Poli, <b>R. Ohana</b>, L. Daudet, I. Carron, M. Filippone. <a href="https://www.eurecom.fr/en/publication/6739" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">ASPAI 2021</a></li>
268
269              <li>  <i>Reservoir Computing meets Recurrent Kernels and Structured Transforms.</i> <b>R. Ohana<sup>*</sup></b>, J. Dong<sup>*</sup>, M. Rafayelyan, F. Krzakala. <a href="https://papers.nips.cc/paper/2020/file/c348616cd8a86ee661c7c98800678fad-Paper.pdf" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt"><u>Oral</u> @ NeurIPS 2020</a>   </li> 
270              
271              <li>  <i>Experimental Approach to Demonstrating Contextuality for Qudits.</i> A. Sohbi, <b>R. Ohana</b>, I. Zaquine, E. Diamanti, D. Markham. <a href="https://arxiv.org/abs/2010.13278" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">Physical Review A</a>   </li> 
272              
273              <li>  <i>Kernel Computations from large-scale random features obtained by Optical Processing Units.</i> <b>R. Ohana</b>, J. Wacker, J. Dong, S. Marmin, F. Krzakala, M. Filippone, L. Daudet. <a href="https://arxiv.org/abs/1910.09880" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">ICASSP 2020</a>   </li> 
274              
275              <li>  <i>Impact of epitaxial strain on the topological-nontopological phase diagram and semimetallic behavior of InAs/GaSb composite quantum wells.</i> H. Irie, T. Akiho, F. Couedo, <b>R. Ohana</b>, S. Suzuki, H. Onomitsu, K. Muraki. <a href="https://arxiv.org/abs/2002.12503v1" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">Physical Review B</a></li> 
276              
277            </ol>
278
279              <h2>Patent</h2> <ul> 
280                
281              <li>  <i>Method and System for machine learning using optical data.</i> I. Poli, J. Launay, K. Müller, G. Pariente, I. Carron, L. Daudet, <b>R. Ohana</b>, D. Hesslow. <a href="https://patentimages.storage.googleapis.com/a1/34/7c/3c91a8598ebe1f/US20210287079A1.pdf" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">US Patent</a>   
282              </li> </ul> 
283              
284              <h2>PhD Manuscript</h2> <ul> 
285                  
286              <li>  <i>Leveraging (physical) randomness in machine learning algorithms.</i> <b>R. Ohana</b>. <a href="https://theses.hal.science/tel-03828849/document" rel="noreferrer noopener" target="_blank" class="button primary icon xsmall solid fa-external-link-alt">PhD Thesis</a>   </li> </ul> 
287              
288              <h2>Miscellaneous</h2> <ul> 
289              <li>Student supervision: 
290                <ul>
291                <li><a href="https://david-heurtel-depeiges.github.io/">David Heurtel-Depeiges </a>(PhD student at University of Montreal).</li>
292                <li><a href="https://people.csail.mit.edu/theoxo/">Theo Olausson</a> (PhD student at MIT).
293                </li>
294                <li> <a href="https://francois-rozet.github.io/"> François Rozet</a> (PhD student at University of Liège).</li>
295                <li> <a href="https://www.linkedin.com/in/mhsotoudeh/"> Mohammad-Hadi Sotoudeh</a> (PhD student at Cambridge University).</li>
296               </ul>
297              </li>
298              <li>Reviewer for NeurIPS 2021-2024, ICML 2022-2024, ICLR 2023, ALT 2020, WACV 2022, Nature Communications, Journal of Machine Learning Research (JMLR).</li> 
299              </ul>
300          </div>
301        </div>
302      </section>
303      <!-- Talks -->
304      <section id="talks">
305        <h2>Talks</h2>
306        <div class="row">
307          <div class="col-12 col-12-small">
308            <!-- TODO: Add talks  -->
309            <ul>
310              <li>
311                Building foundation models for science, Brookhaven National Lab 2024.
312              </li>
313              <li>
314                Building foundation models for science, Sorbonne University 2024.
315              </li>
316              <li>
317                Building foundation models for science, M2Lines annual meeting 2024.
318              </li>
319              <li>
320                <a href="https://www.youtube.com/watch?v=_sKZsx7Iprg"
321                  >100 seconds Research Challenge on Gibbs diffusion</a
322                >, ICML 2024 by Neptune AI.
323              </li>
324              <li>
325                <a href="https://simons.berkeley.edu/workshops/aiscience-strengthening-bond-between-sciences-artificial-intelligence">AI for Science Workshop 2024</a>, UC Berkeley.
326              </li>
327              <li>Building Foundation models for science at SuperComputing 2023.</li>
328
329              <li>
330                <a href="https://nips.cc/virtual/2020/public/session_oral_21072.html"
331                  >Oral @ NeurIPS 2020 (starting at 46:30).</a
332                >
333              </li>
334              <li>
335                Paris Machine Learning Meetup 2020.
336              </li>
337              <li>
338                Dauphine Journal club 2020.
339              </li>
340              <li>
341                Roscoff meeting 2019.
342              </li>
343            </ul>
344          </div>
345        </div>
346      </section>
347      <!-- Software -->
348      <!-- <section id="software">
349        <h2>Software</h2>
350        TODO: Add Software  -->
351        <!-- <p>I contribute to a number of open source software projects.</p>
352        <div class="table-wrapper">
353          <table class="alt">
354            <thead>
355              <tr>
356                <th>
356Name</th>
357                <th>Description</th>
358              </tr>
359            </thead>
360            <tbody>
361              <tr>
362                <td><a href="#">Item One</a></td>
363                <td>
364                  Lorem ipsum dolor sit amet, consectetur adipiscing elit.
365                </td>
366              </tr>
367              <tr>
368                <td><a href="#">Item Two</a></td>
369                <td>
370                  Vis sunt in culpa qui officia deserunt mollit anim id est
371                  laborum.
372                </td>
373              </tr>
374              <tr>
375                <td><a href="#">Item Three</a></td>
376                <td>
377                  Ut sed do eiusmod tempor incididunt ut labore et dolore magna
378                  aliqua.
379                </td>
380              </tr>
381              <tr>
382                <td><a href="#">Item Four</a></td>
383                <td>
384                  Ut enim ad minim veniam, quis nostrud exercitation ullamco
385                  laboris.
386                </td>
387              </tr>
388              <tr>
389                <td><a href="#">Item Five</a></td>
390                <td>
391                  Duis aute irure dolor in reprehenderit in voluptate velit.
392                </td>
393              </tr>
394            </tbody>
395          </table>
396        </div>
397      </section> --> 
398      <!-- Contact -->
399      <section id="contact">
400        <h2>Contact</h2>
401        <div class="row">
402          <div class="col-6 col-12-small">
403            <ul class="icons">
404              <li>
405                <a href="https://www.nvidia.com/en-us/" class="icon solid fa-home">
406                  <span class="label">Address</span>
407                </a>
408                NVIDIA Corporation<br />
409                2788 San Tomas Expy<br />
410                Santa Clara, CA 95051<br />
411                United States<br />
412                <a href="mailto:[email protected]" class="icon solid fa-envelope">
413                  <span class="label">Email</span>
414                </a>
415                <a href="https://twitter.com/oharub" class="icon brands fa-twitter">
416                  <span class="label">Twitter</span>
417                </a>
418                <a href="https://www.linkedin.com/in/rubenohana/" class="icon brands fa-linkedin">
419                  <span class="label">LinkedIn</span>
420                </a>
421              </li>
422            </ul>
423          </div>
424        </div>
425      </section>
426    </div>
427    <!-- Footer -->
428    <footer id="footer">
429      <div class="inner" style="text-align: right; padding-right: 0em;">
430        <a
431          href="https://www.nvidia.com"
432          class="image logo"
433          target="_blank"
434          id="logo"
435          rel="noreferrer noopener"
436          style="float: right; margin-left: 2em;"
437        >
438          <img src="images/nvidia_logo_white_oldtodelete.svg" alt="NVIDIA logo" id="logo-icon" style="border-radius: 0;" />
439        </a>
440      </div>
441    </footer>
442    <!-- Scripts -->
443    
443<script src="assets/js/jquery.min.js"></script>
443
444    
444<script src="assets/js/jquery.poptrox.min.js"></script>
444
445    
445<script src="assets/js/jquery.scrollex.min.js"></script>
445
446    
446<script src="assets/js/jquery.scrolly.min.js"></script>
446
447    
447<script src="assets/js/browser.min.js"></script>
447
448    
448<script src="assets/js/breakpoints.min.js"></script>
448
449    
449<script src="assets/js/util.js"></script>
449
450    
450<script src="assets/js/main.js"></script>
450
451  </body>
452</html>

Line numbers count LF bytes from the start of the resource, as the search results do. Vendor segments are library code the classifier recognised; they are stored but not indexed. Bytes are shown as Latin1 characters, one per byte.