Vincent Roulet

Stochastic Optimization for Spectral Risk Measures

Dec 10, 2022
Ronak Mehta, Vincent Roulet, Krishna Pillutla, Lang Liu, Zaid Harchaoui

Figure 2 for Stochastic Optimization for Spectral Risk Measures

Figure 3 for Stochastic Optimization for Spectral Risk Measures

Figure 4 for Stochastic Optimization for Spectral Risk Measures

Access Paper or Ask Questions

Iterative Linear Quadratic Optimization for Nonlinear Control: Differentiable Programming Algorithmic Templates

Jul 13, 2022
Vincent Roulet, Siddhartha Srinivasa, Maryam Fazel, Zaid Harchaoui

Figure 2 for Iterative Linear Quadratic Optimization for Nonlinear Control: Differentiable Programming Algorithmic Templates

Figure 3 for Iterative Linear Quadratic Optimization for Nonlinear Control: Differentiable Programming Algorithmic Templates

Figure 4 for Iterative Linear Quadratic Optimization for Nonlinear Control: Differentiable Programming Algorithmic Templates

* This is a companion report to the arXiv report "Complexity Bounds of Iterative Linear Quadratic Optimization Algorithms for Discrete Time Nonlinear Control" by the same authors

Access Paper or Ask Questions

Target Propagation via Regularized Inversion

Dec 02, 2021
Vincent Roulet, Zaid Harchaoui

Figure 2 for Target Propagation via Regularized Inversion

Figure 3 for Target Propagation via Regularized Inversion

Figure 4 for Target Propagation via Regularized Inversion

Access Paper or Ask Questions

Differentiable Programming à la Moreau

Dec 31, 2020
Vincent Roulet, Zaid Harchaoui

Figure 2 for Differentiable Programming à la Moreau

Figure 3 for Differentiable Programming à la Moreau

Figure 4 for Differentiable Programming à la Moreau

Access Paper or Ask Questions

An Elementary Approach to Convergence Guarantees of Optimization Algorithms for Deep Networks

Feb 20, 2020
Vincent Roulet, Zaid Harchaoui

* Short version appeared in the proceedings of the 57th Annual Allerton Conference on Communication, Control, and Computing (Allerton 2019)

Access Paper or Ask Questions

End-to-end Learning, with or without Labels

Dec 30, 2019
Corinne Jones, Vincent Roulet, Zaid Harchaoui

Figure 2 for End-to-end Learning, with or without Labels

Figure 3 for End-to-end Learning, with or without Labels

Figure 4 for End-to-end Learning, with or without Labels

Access Paper or Ask Questions

Kernel-based Translations of Convolutional Networks

Mar 19, 2019
Corinne Jones, Vincent Roulet, Zaid Harchaoui

Figure 2 for Kernel-based Translations of Convolutional Networks

Figure 3 for Kernel-based Translations of Convolutional Networks

Figure 4 for Kernel-based Translations of Convolutional Networks

Access Paper or Ask Questions

A Smoother Way to Train Structured Prediction Models

Feb 08, 2019
Krishna Pillutla, Vincent Roulet, Sham M. Kakade, Zaid Harchaoui

Figure 2 for A Smoother Way to Train Structured Prediction Models

Figure 3 for A Smoother Way to Train Structured Prediction Models

Figure 4 for A Smoother Way to Train Structured Prediction Models

* Short version appeared in Neural Information Processing Systems (NeurIPS) 2018

Access Paper or Ask Questions

Learning with Clustering Structure

Sep 19, 2016
Vincent Roulet, Fajwel Fogel, Alexandre d'Aspremont, Francis Bach

Figure 2 for Learning with Clustering Structure

Figure 3 for Learning with Clustering Structure

Figure 4 for Learning with Clustering Structure

* Completely rewritten. New convergence proofs in the clustered and sparse clustered case. New projection algorithm on sparse clustered vectors

Access Paper or Ask Questions