Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox
Picture for Volkan Cevher

Volkan Cevher

On the Complexity of a Practical Primal-Dual Coordinate Method


Jan 19, 2022
Ahmet Alacaoglu, Volkan Cevher, Stephen J. Wright


  Access Paper or Ask Questions

Subquadratic Overparameterization for Shallow Neural Networks


Nov 02, 2021
Chaehwan Song, Ali Ramezani-Kebrya, Thomas Pethick, Armin Eftekhari, Volkan Cevher

Figure 1 for Subquadratic Overparameterization for Shallow Neural Networks
Figure 2 for Subquadratic Overparameterization for Shallow Neural Networks
* To appear at the conference on Neural Information Processing Systems (NeurIPS 2021) 
  Access Paper or Ask Questions

STORM+: Fully Adaptive SGD with Momentum for Nonconvex Optimization


Nov 01, 2021
Kfir Y. Levy, Ali Kavis, Volkan Cevher

Figure 1 for STORM+: Fully Adaptive SGD with Momentum for Nonconvex Optimization
* 25 pages, 1 figure, accepted to NeurIPS 2021 
  Access Paper or Ask Questions

On the Double Descent of Random Features Models Trained with SGD


Oct 29, 2021
Fanghui Liu, Johan A. K. Suykens, Volkan Cevher

Figure 1 for On the Double Descent of Random Features Models Trained with SGD
Figure 2 for On the Double Descent of Random Features Models Trained with SGD
Figure 3 for On the Double Descent of Random Features Models Trained with SGD
* 34 pages, 2 figures. This version gives a refined estimation on B2 (one term in Bias) 
  Access Paper or Ask Questions

A first-order primal-dual method with adaptivity to local smoothness


Oct 28, 2021
Maria-Luiza Vladarean, Yura Malitsky, Volkan Cevher

Figure 1 for A first-order primal-dual method with adaptivity to local smoothness
Figure 2 for A first-order primal-dual method with adaptivity to local smoothness
Figure 3 for A first-order primal-dual method with adaptivity to local smoothness

  Access Paper or Ask Questions

Self-Supervised Neural Architecture Search for Imbalanced Datasets


Sep 20, 2021
Aleksandr Timofeev, Grigorios G. Chrysos, Volkan Cevher

Figure 1 for Self-Supervised Neural Architecture Search for Imbalanced Datasets
Figure 2 for Self-Supervised Neural Architecture Search for Imbalanced Datasets
Figure 3 for Self-Supervised Neural Architecture Search for Imbalanced Datasets
* Published in ICML 2021 Workshop: Self-Supervised Learning for Reasoning and Perception. Code: https://github.com/TimofeevAlex/ssnas_imbalanced 
  Access Paper or Ask Questions

Regret minimization in stochastic non-convex learning via a proximal-gradient approach


Oct 13, 2020
Nadav Hallak, Panayotis Mertikopoulos, Volkan Cevher


  Access Paper or Ask Questions

Efficient Proximal Mapping of the 1-path-norm of Shallow Networks


Jul 15, 2020
Fabian Latorre, Paul Rolland, Nadav Hallak, Volkan Cevher

Figure 1 for Efficient Proximal Mapping of the 1-path-norm of Shallow Networks
Figure 2 for Efficient Proximal Mapping of the 1-path-norm of Shallow Networks
Figure 3 for Efficient Proximal Mapping of the 1-path-norm of Shallow Networks
Figure 4 for Efficient Proximal Mapping of the 1-path-norm of Shallow Networks
* ICML 2020. Fabian Latorre, Paul Rolland and Nadav Hallak have contributed equally 
  Access Paper or Ask Questions

Random extrapolation for primal-dual coordinate descent


Jul 13, 2020
Ahmet Alacaoglu, Olivier Fercoq, Volkan Cevher

Figure 1 for Random extrapolation for primal-dual coordinate descent
Figure 2 for Random extrapolation for primal-dual coordinate descent
Figure 3 for Random extrapolation for primal-dual coordinate descent
Figure 4 for Random extrapolation for primal-dual coordinate descent
* To appear in ICML 2020 
  Access Paper or Ask Questions

Conditional gradient methods for stochastically constrained convex minimization


Jul 07, 2020
Maria-Luiza Vladarean, Ahmet Alacaoglu, Ya-Ping Hsieh, Volkan Cevher

Figure 1 for Conditional gradient methods for stochastically constrained convex minimization
Figure 2 for Conditional gradient methods for stochastically constrained convex minimization
Figure 3 for Conditional gradient methods for stochastically constrained convex minimization
Figure 4 for Conditional gradient methods for stochastically constrained convex minimization

  Access Paper or Ask Questions

Robust Inverse Reinforcement Learning under Transition Dynamics Mismatch


Jul 02, 2020
Luca Viano, Yu-Ting Huang, Parameswaran Kamalaruban, Volkan Cevher

Figure 1 for Robust Inverse Reinforcement Learning under Transition Dynamics Mismatch
Figure 2 for Robust Inverse Reinforcement Learning under Transition Dynamics Mismatch
Figure 3 for Robust Inverse Reinforcement Learning under Transition Dynamics Mismatch
Figure 4 for Robust Inverse Reinforcement Learning under Transition Dynamics Mismatch

  Access Paper or Ask Questions

Interaction-limited Inverse Reinforcement Learning


Jul 01, 2020
Martin Troussard, Emmanuel Pignat, Parameswaran Kamalaruban, Sylvain Calinon, Volkan Cevher

Figure 1 for Interaction-limited Inverse Reinforcement Learning
Figure 2 for Interaction-limited Inverse Reinforcement Learning
Figure 3 for Interaction-limited Inverse Reinforcement Learning
Figure 4 for Interaction-limited Inverse Reinforcement Learning

  Access Paper or Ask Questions

Environment Shaping in Reinforcement Learning using State Abstraction


Jun 23, 2020
Parameswaran Kamalaruban, Rati Devidze, Volkan Cevher, Adish Singla

Figure 1 for Environment Shaping in Reinforcement Learning using State Abstraction
Figure 2 for Environment Shaping in Reinforcement Learning using State Abstraction
Figure 3 for Environment Shaping in Reinforcement Learning using State Abstraction

  Access Paper or Ask Questions

On the Almost Sure Convergence of Stochastic Gradient Descent in Non-Convex Problems


Jun 19, 2020
Panayotis Mertikopoulos, Nadav Hallak, Ali Kavis, Volkan Cevher

Figure 1 for On the Almost Sure Convergence of Stochastic Gradient Descent in Non-Convex Problems
Figure 2 for On the Almost Sure Convergence of Stochastic Gradient Descent in Non-Convex Problems
Figure 3 for On the Almost Sure Convergence of Stochastic Gradient Descent in Non-Convex Problems
* 32 pages, 8 figures 
  Access Paper or Ask Questions

The limits of min-max optimization algorithms: convergence to spurious non-critical sets


Jun 16, 2020
Ya-Ping Hsieh, Panayotis Mertikopoulos, Volkan Cevher

Figure 1 for The limits of min-max optimization algorithms: convergence to spurious non-critical sets
Figure 2 for The limits of min-max optimization algorithms: convergence to spurious non-critical sets
Figure 3 for The limits of min-max optimization algorithms: convergence to spurious non-critical sets
Figure 4 for The limits of min-max optimization algorithms: convergence to spurious non-critical sets

  Access Paper or Ask Questions

Convergence of adaptive algorithms for weakly convex constrained optimization


Jun 11, 2020
Ahmet Alacaoglu, Yura Malitsky, Volkan Cevher

Figure 1 for Convergence of adaptive algorithms for weakly convex constrained optimization

  Access Paper or Ask Questions

Lipschitz constant estimation of Neural Networks via sparse polynomial optimization


Apr 18, 2020
Fabian Latorre, Paul Rolland, Volkan Cevher

Figure 1 for Lipschitz constant estimation of Neural Networks via sparse polynomial optimization
Figure 2 for Lipschitz constant estimation of Neural Networks via sparse polynomial optimization
Figure 3 for Lipschitz constant estimation of Neural Networks via sparse polynomial optimization
Figure 4 for Lipschitz constant estimation of Neural Networks via sparse polynomial optimization
* Published as a conference paper in ICLR2020, originally submitted in September 25 2019 and available at https://openreview.net/forum?id=rJe4_xSFDB 
  Access Paper or Ask Questions

A new regret analysis for Adam-type algorithms


Mar 21, 2020
Ahmet Alacaoglu, Yura Malitsky, Panayotis Mertikopoulos, Volkan Cevher


  Access Paper or Ask Questions

A Newton Frank-Wolfe Method for Constrained Self-Concordant Minimization


Feb 17, 2020
Deyi Liu, Volkan Cevher, Quoc Tran-Dinh

Figure 1 for A Newton Frank-Wolfe Method for Constrained Self-Concordant Minimization
Figure 2 for A Newton Frank-Wolfe Method for Constrained Self-Concordant Minimization
Figure 3 for A Newton Frank-Wolfe Method for Constrained Self-Concordant Minimization
Figure 4 for A Newton Frank-Wolfe Method for Constrained Self-Concordant Minimization

  Access Paper or Ask Questions

Robust Reinforcement Learning via Adversarial training with Langevin Dynamics


Feb 14, 2020
Parameswaran Kamalaruban, Yu-Ting Huang, Ya-Ping Hsieh, Paul Rolland, Cheng Shi, Volkan Cevher

Figure 1 for Robust Reinforcement Learning via Adversarial training with Langevin Dynamics
Figure 2 for Robust Reinforcement Learning via Adversarial training with Langevin Dynamics
Figure 3 for Robust Reinforcement Learning via Adversarial training with Langevin Dynamics
Figure 4 for Robust Reinforcement Learning via Adversarial training with Langevin Dynamics

  Access Paper or Ask Questions

Optimization for Reinforcement Learning: From Single Agent to Cooperative Agents


Dec 01, 2019
Donghwan Lee, Niao He, Parameswaran Kamalaruban, Volkan Cevher

Figure 1 for Optimization for Reinforcement Learning: From Single Agent to Cooperative Agents
Figure 2 for Optimization for Reinforcement Learning: From Single Agent to Cooperative Agents
Figure 3 for Optimization for Reinforcement Learning: From Single Agent to Cooperative Agents
Figure 4 for Optimization for Reinforcement Learning: From Single Agent to Cooperative Agents

  Access Paper or Ask Questions

UniXGrad: A Universal, Adaptive Algorithm with Optimal Guarantees for Constrained Optimization


Oct 30, 2019
Ali Kavis, Kfir Y. Levy, Francis Bach, Volkan Cevher

Figure 1 for UniXGrad: A Universal, Adaptive Algorithm with Optimal Guarantees for Constrained Optimization
Figure 2 for UniXGrad: A Universal, Adaptive Algorithm with Optimal Guarantees for Constrained Optimization
Figure 3 for UniXGrad: A Universal, Adaptive Algorithm with Optimal Guarantees for Constrained Optimization
* NeurIPS 2019 
  Access Paper or Ask Questions

Nearly Minimal Over-Parametrization of Shallow Neural Networks


Oct 29, 2019
Armin Eftekhari, ChaeHwan Song, Volkan Cevher

* This paper is submitted without consent of the co-authors 
  Access Paper or Ask Questions

Fast and Provable ADMM for Learning with Generative Priors


Jul 07, 2019
Fabian Latorre Gómez, Armin Eftekhari, Volkan Cevher

Figure 1 for Fast and Provable ADMM for Learning with Generative Priors
Figure 2 for Fast and Provable ADMM for Learning with Generative Priors
Figure 3 for Fast and Provable ADMM for Learning with Generative Priors
Figure 4 for Fast and Provable ADMM for Learning with Generative Priors

  Access Paper or Ask Questions

Interactive Teaching Algorithms for Inverse Reinforcement Learning


Jun 05, 2019
Parameswaran Kamalaruban, Rati Devidze, Volkan Cevher, Adish Singla

Figure 1 for Interactive Teaching Algorithms for Inverse Reinforcement Learning
Figure 2 for Interactive Teaching Algorithms for Inverse Reinforcement Learning
Figure 3 for Interactive Teaching Algorithms for Inverse Reinforcement Learning
Figure 4 for Interactive Teaching Algorithms for Inverse Reinforcement Learning
* IJCAI'19 paper (extended version) 
  Access Paper or Ask Questions