Vincent Roulet
Scholar

Vincent Roulet

Google Scholar ID: vwoZrVMAAAAJ
Google DeepMind
OptimizationNonlinear ControlMachine Learning
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
996
 
H-index
9
 
i10-index
9
 
Publications
20
 
Co-authors
0
 
Contact
GitHubOpen ↗
Publications
7 items
Autoregressive Language Models are Secretly Energy-Based Models: Insights into the Lookahead Capabilities of Next-Token Prediction
2025
Cited
0
Per-example gradients: a new frontier for understanding and improving optimizers
2025
Cited
0
How far away are truly hyperparameter-free learning algorithms?
2025
Cited
0
Loss Functions and Operators Generated by f-Divergences
2025
Cited
0
Joint Learning of Energy-based Models and their Partition Function
2025
Cited
0
The Elements of Differentiable Programming
arXiv.org · 2024
Cited
40
Iterative Linear Quadratic Optimization for Nonlinear Control: Differentiable Programming Algorithmic Templates
arXiv.org · 2022
Cited
6
Resume
Academic Achievements
  • No specific academic achievements mentioned.
Research Experience
  • Currently a Research Scientist at Google DeepMind. Previously, an acting assistant professor in the Department of Statistics at the University of Washington, and before that a postdoctoral fellow in the NSF-TRIPODS institute ADSI (now IFDS) at the University of Washington, working with Zaid Harchaoui.
Education
  • Ph.D. from Ecole Normale Superieure de Paris, supervised by Alexandre d’Aspremont, within the Sierra Team led by Francis Bach.
Background
  • Optimization and Machine Learning Researcher. Currently interested in developing new algorithms and convergence guarantees for optimization algorithms in a differentiable programming framework.
Miscellany
  • No other information mentioned.