Carles Domingo-Enrich

I am a fourth year PhD student in Computer Science at the Courant Institute of Mathematical Sciences (New York University). My advisor is Prof. Joan Bruna. I am working on theoretical aspects of generative modeling, theory of deep learning, optimization for continuous games, and related topics at the intersection of machine learning, statistics and optimization. In 2019 I obtained a B.S. in Mathematics and a B.S. in Engineering Physics from the Polytechnic University of Catalonia (UPC). In the summer of 2021 I interned at IBM Research AI (Yorktown Heights, NY) with Youssef Mroueh. In the summer of 2022 I interned at Microsoft Research New England (Cambridge, MA) with Lester Mackey. During the 2022-2023 school year I am a Visiting Researcher (part-time) at Meta in New York City, working with Ricky T. Q. Chen and Brandon Amos. [My CV] [My GitHub page] [My Google Scholar page]

Preprints:

Computing the variance of shuffling stochastic gradient algorithms via power spectral density analysis
Carles Domingo-Enrich
Jun. 2022 [PDF]
Auditing differential privacy in high dimensions with the kernel quantum Rényi divergence
Carles Domingo-Enrich, Youssef Mroueh
May 2022 [PDF]
Simultaneous transport evolution for minimax equilibria on measures
Carles Domingo-Enrich, Joan Bruna
Feb. 2022 [PDF]
Dual training of energy-based models with overparametrized shallow neural networks
Carles Domingo-Enrich, Alberto Bietti, Marylou Gabrié, Joan Bruna, Eric Vanden-Eijnden
July 2021. [PDF]

Publications:

An explicit expansion of the Kullback-Leibler divergence along its Fisher-Rao gradient flow
Carles Domingo-Enrich, Aram-Alexandre Pooladian
TMLR 2023. [PDF]
Multisample Flow Matching: Straightening Flows with Minibatch Couplings
Aram-Alexandre Pooladian*, Heli Ben-Hamu*, Carles Domingo-Enrich*, Brandon Amos, Yaron Lipman, Ricky Chen (*Equal contribution)
ICML 2023. [PDF]
Compress then test: powerful kernel testing in near-linear time
Carles Domingo-Enrich, Raaz Dwivedi, Lester Mackey
AISTATS 2023. [PDF]
Learning with stochastic orders
Carles Domingo-Enrich, Yair Schiff, Youssef Mroueh
ICLR 2023. Spotlight. [PDF]
Depth and feature learning are provably beneficial for neural network discriminators
Carles Domingo-Enrich
COLT 2022. [PDF]
Tighter sparse approximation bounds for ReLU neural networks
Carles Domingo-Enrich, Youssef Mroueh
ICLR 2022. Spotlight. [PDF]
Separation results between fixed-kernel and feature-learning probability metrics
Carles Domingo-Enrich, Youssef Mroueh
NeurIPS 2021. Oral. [PDF]
On energy-based models with overparametrized shallow neural networks
Carles Domingo-Enrich, Alberto Bietti, Eric Vanden-Eijnden, Joan Bruna
ICML 2021. Long talk. [PDF]
Average-case acceleration for bilinear games and normal matrices
Carles Domingo-Enrich, Damien Scieur, Fabian Pedregosa
ICLR 2021 [PDF]
A mean-field analysis of two-player zero-sum games
Carles Domingo-Enrich, Samy Jelassi, Arthur Mensch, Grant M. Rotskoff and Joan Bruna
NeurIPS 2020 [PDF]
Extra-gradient with player sampling for faster Nash equilibrium finding
Samy Jelassi*, Carles Domingo-Enrich*, Damien Scieur, Arthur Mensch and Joan Bruna (*Equal contribution)
ICML 2020 [PDF]
Outsourcing scalar products and matrix products on privacy-protected unencrypted data stored in untrusted clouds
Josep Domingo-Ferrer, Sara Ricci and Carles Domingo-Enrich
Information Sciences, Vol. 436, pp. 320-342, Apr 2018 [DOI] [PDF]

Old courses

During Fall 2021, I was the section leader for the two sections of DS-GA 1005: Inference and Representation. Section 2 meets on Mondays from 7.00 to 7.50 pm at the Global Center for Academic & Spiritual Life, 238 Thompson Street, room 261. Section 3 meets on Mondays from 4.00 to 4.50 pm at the Silver Center for Arts & Science, 100 Washington Square East, room 520. At least the first week, the classes will be recorded on Zoom (see Brightspace page of the course for link). Office hours are on Mondays from 5.00 to 6.45 pm at 60 5th Avenue, room 204 and on Zoom at the same time (see Brightspace page of the course for link).

Notes for the recitations:

During Fall 2020, I was the remote section leader for Section 2 and Section 4 of DS-GA 1014: Optimization and Computational Linear Algebra for Data Science.

Slides for remote recitations (with me):

Other material for that course available in this folder.

Address

New York University, Courant Institute of Mathematical Sciences
Department of Computer Science
Office 550
60 Fifth Avenue
New York, NY 10011

email: cd2754 (at) nyu.edu