| CARVIEW |
Select Language
HTTP/2 200
server: GitHub.com
content-type: text/html; charset=utf-8
last-modified: Sun, 10 Sep 2023 21:49:36 GMT
access-control-allow-origin: *
strict-transport-security: max-age=31556952
etag: W/"64fe39f0-7286"
expires: Mon, 29 Dec 2025 08:29:10 GMT
cache-control: max-age=600
content-encoding: gzip
x-proxy-cache: MISS
x-github-request-id: B3EA:3A7A40:87B1DA:98833C:6952397E
accept-ranges: bytes
age: 0
date: Mon, 29 Dec 2025 08:19:11 GMT
via: 1.1 varnish
x-served-by: cache-bom-vanm7210037-BOM
x-cache: MISS
x-cache-hits: 0
x-timer: S1766996351.851796,VS0,VE205
vary: Accept-Encoding
x-fastly-request-id: 12fc46e87959a0fa1b6606cebc1c7d1f87929d14
content-length: 6796
vdb
Valentin De Bortoli
Biography
I am a research scientist at Google DeepMind in London, working with Arnaud
Doucet. A list of my publications with filters is
available here and here
is
my Google
Scholar. See also my CV (English).
Contact:
vdebortoli@google.com
For those interested in diffusion models, with James Thornton we have created a website listing papers on diffusion models. Please point out any missing reference!
Interests
- Machine Learning
- Information geometry
- Markov chain theory
- Optimal transport
Short CV
-
Research scientist, 2023 - now
Google DeepMind
-
CNRS researcher (on leave), 2022 - now
ENS Ulm
-
Postdoc, 2020 - 2022
Oxford University
-
PhD, 2017 - 2020
ENS Paris Saclay
Phillips, Seror, Hutchinson, De Bortoli, Doucet, Mathieu.
Spectral Diffusion Processes.
2022 - Submitted (conference).
Crucinio, De Bortoli, Doucet, Johansen.
Solving Fredholm Integral Equations of the First Kind via Wasserstein Gradient Flows.
2022 - Submitted (journal).
De Bortoli.
Convergence of denoising diffusion models under the manifold hypothesis.
2022 - Accepted at TMLR (journal).
Thornton, Hutchinson, Mathieu, De Bortoli, Teh, Doucet.
Riemannian Diffusion Schrödinger Bridges.
2022 - Accepted at ICML (conference).
Shi, De Bortoli, Deligiannidis, Doucet.
Conditional Simulation Using Diffusion Schrödinger Bridges.
2022 - Accepted at UAI (conference).
Campbell, Benton, De Bortoli, Rainforth, Deligiannidis, Doucet.
A Continuous Time Framework for Discrete Denoising Models.
2022 - Accepted at Neurips (conference).
Salmona, Delon, Desolneux, De Bortoli.
Can Push-forward Generative Models Fit Multimodal Distributions.
2022 - Accepted at Neurips (conference).
Guth, Coste, De Bortoli, Mallat.
Wavelet Score-Based Generative Modeling.
2022 - Accepted at Neurips (conference).
De Bortoli, Mathieu, Hutchinson, Thornton, Teh, Doucet.
Riemannian Score Based Generative Modeling.
2022 - Accepted at Neurips (conference).
Laumont, De Bortoli, Almansa, Delon, Durmus, Pereyra.
Maximum-a-Posteriori estimation with Plug & Play priors and stochastic
gradient descent
.
2022 - Accepted at JMIV (journal).
De Bortoli, Desolneux.
On quantitative Laplace-type convergence results for some exponential probability measures, with two applications
.
2021 - Accepted at JMLR (journal).
De Bortoli, Doucet, Heng, Thornton.
Simulating Diffusion Bridges with Score Matching
.
2021 - Submitted (journal)
Deligiannidis, De Bortoli, Doucet.
Quantitative Uniform Stability of the Iterative Proportional Fitting Procedure
.
2021 - Submitted (journal)
De Bortoli, Thornton, Heng, Doucet.
Diffusion Schrödinger Bridge with Applications to Score-Based Generative Modeling.
2021 - Accepted at NeuRIPS (conference).
Laumont, De Bortoli, Almansa, Delon, Durmus, Pereyra.
Bayesian imaging using Plug & Play priors: when Langevin meets Tweedie.
2021 - Accepted at SIAM Imaging Science(journal).
De Bortoli, Desolneux, Durmus, Galerne, Leclaire.
Maximum entropy methods for texture synthesis: theory and practice.
2021 - Published in SIMODS (journal).
De Bortoli, Durmus, Fontaine, Simsekli.
Quantitative
Propagation of Chaos for SGD in Wide Neural Networks.
2020 - Accepted at NeuRIPS (conference).
F. Vidal, De Bortoli, Pereyra, Durmus.
Maximum likelihood estimation of regularisation parameters in
high-dimensional inverse problems: an empirical Bayesian. Part I: Methodology and Experiments
approach.
2020 - Published in SIIMS (journal).
De Bortoli, Durmus, F.Vidal, Pereyra.
Maximum likelihood estimation of regularisation parameters in
high-dimensional inverse problems: an empirical Bayesian. Part II: Theoretical Analysis
approach.
2020 - Published in SIIMS (journal).
Nadjahi, De Bortoli, Durmus, Badeau, Simsekli.
Approximate Bayesian Computation with the Sliced-Wasserstein Distance.
2020 - Accepted at ICASSP (conference) (Best student paper award).
De Bortoli, Desolneux, Galerne, Leclaire.
Redundancy in Gaussian random fields.
2020 - Published in ESAIM: Probability and Statistics
Fontaine, De Bortoli, Durmus.
Continuous and Discrete-Time Analysis of Stochastic Gradient Descent for Convex and Non-Convex Functions.
2021 - Accepted at COLT (conference).
De Bortoli, Desolneux, Galerne, Leclaire.
Patch redundancy in images: a statistical testing framework and some
applications.
2019 - Published in SIAM Journal on Imaging Science (journal).
De Bortoli, Durmus, Pereyra, F.Vidal.
Efficient stochastic optimisation by unadjusted Langevin Monte Carlo. Application to maximum marginal likelihood and empirical Bayesian estimation.
2019 - Accepted at Statistics and Computing (journal).
De Bortoli, Durmus.
Convergence of diffusion and their discretizations: from continuous to discrete processes and back.
2019 - Submitted (journal).
De Bortoli, Desolneux, Galerne, Leclaire.
Macrocanonical models for texture synthesis.
2019 - Accepted at Scale Space and Variational Methods in Computer Vision (conference).
Huang, De Bortoli, Zhou, Gilles.
Review of wavelet-based unsupervised texture segmentation, advantage of adaptive wavelets.
2018 - Published in IET (journal).
Teaching
.2017-2018
Generative Modeling
MVA course (M2) generative modeling
Differential Calculus
ENS Paris Saclay L3 differential calculus
Hilbert Analysis
ENS Paris Saclay agrégation Hilbert analysis
Fourier Analysis
ENS Paris Saclay agrégation Fourier analysis
Optimization
ENS Paris Saclay M1 optimization
Contact
- vdebortoli@google.com
- 29 St Giles', Oxford OX1, UK