Volkan Cevher
Associate Professor
EPFL STI IEL LIONS
ELE 233 (Bâtiment ELE)
Station 11
1015 Lausanne
+41 21 693 11 01
+41 21 693 11 74
Office:
ELE 233
EPFL
>
STI
>
IEM
>
LIONS
Web site: Web site: https://lions.epfl.ch
+41 21 693 11 01
EPFL
>
STI
>
STI-SEL
>
SEL-ENS
+41 21 693 11 01
EPFL
>
P
>
P-SG
>
CCE
Web site: Web site: https://cce.epfl.ch/
Fields of expertise
Machine Learning
Optimization
Signal Processing
Information Theory
Optimization
Signal Processing
Information Theory
Biography
Volkan Cevher received the B.Sc. (valedictorian) in electrical engineering from Bilkent University in Ankara, Turkey, in 1999 and the Ph.D. in electrical and computer engineering from the Georgia Institute of Technology in Atlanta, GA in 2005. He was a Research Scientist with the University of Maryland, College Park from 2006-2007 and also with Rice University in Houston, TX, from 2008-2009. Currently, he is an Associate Professor at the Swiss Federal Institute of Technology Lausanne and a Faculty Fellow in the Electrical and Computer Engineering Department at Rice University. His research interests include machine learning, signal processing theory, optimization theory and methods, and information theory. Dr. Cevher is an ELLIS fellow and was the recipient of the Google Faculty Research award in 2018, the IEEE Signal Processing Society Best Paper Award in 2016, a Best Paper Award at CAMSAP in 2015, a Best Paper Award at SPARS in 2009, and an ERC CG in 2016 as well as an ERC StG in 2011.Publications
Infoscience publications
Selected Publications
Stable Nonconvex-Nonconcave Training via Linear Interpolation
2023-09-21. Thirty-seventh Conference on Neural Information Processing Systems, New Orleans, Louisiana, USA, December 10-16, 2023.Semi Bandit Dynamics in Congestion Games: Convergence to Nash Equilibrium and No-Regret Guarantees.
2023. 40th International Conference on Machine Learning (ICML), Honolulu, Hawaii, USA, July, 23-29, 2023.Benign Overfitting in Deep Neural Networks under Lazy Training
2023. 40th International Conference on Machine Learning (ICML), Honolulu, Hawaii, USA, July, 23-29, 2023.What can online reinforcement learning with function approximation benefitfrom general coverage conditions
2023. 40th International Conference on Machine Learning (ICML), Honolulu, Hawaii, USA, July, 23-29, 2023.Universal and adaptive methods for robust stochastic optimization
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-9077.Multi-agent Learning with Privacy Guarantees
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-9939.Robust Training and Verification of Deep Neural Networks
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-9223.Text Representation Learning for Low Cost Natural Language Understanding
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-9913.Adversarial Training Should Be Cast As a Non-Zero-Sum Game
2023When do Minimax-fair Learning and Empirical Risk Minimization Coincide?
2023. 40th International Conference on Machine Learning (ICML), Honolulu, Hawaii, USA, July 23-29, 2023.Federated Learning under Covariate Shifts with Generalization Guarantees
Transactions on Machine Learning Research. 2023.Augmented Lagrangian Methods for Provable and Scalable Machine Learning
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-9613.Regularization of polynomial networks for image recognition
2023. Computer Vision and Pattern Recognition Conference (CVPR), Vancouver, Canada, 18-22 June, 2023.Improving Generalization of Pretrained Language Models
Lausanne, EPFL, 2023. DOI : 10.5075/epfl-thesis-8664.Distributed Extra-Gradient With Optimal Complexity And Communication Guarantees
2023. 11th International Conference on Learning Representations (ICLR), Kigali, Rwanda, May 1-5, 2023.Finding Actual Descent Directions For Adversarial Training
2023. 11th International Conference on Learning Representations (ICLR), Kigali, Rwanda, May 1-5, 2023.DiGress: Discrete Denoising diffusion for graph generation
2023. 11th International Conference on Learning Representations (ICLR), Kigali, Rwanda, May 1-5, 2023.Solving stochastic weak Minty variational inequalities without increasing batch size
11th International Conference on Learning Representations ICLR2023, Kigali, Rwanda, May 1-5, 2023.Revisiting adversarial training for the worst-performing class
Transactions on Machine Learning Research. 2023.A Natural Actor-Critic Framework for Zero-Sum Markov Games
2022. 39th International Conference on Machine Learning (ICML), Baltimore, Maryland, USA, July 17-23, 2022.UNDERGRAD: A Universal Black-Box Optimization Method with Almost Dimension-Free Convergence Rate Guarantees
2022. 39th International Conference on Machine Learning (ICML), Baltimore, Maryland, USA, July 17-23, 2022.No-regret learning in games with noisy feedback: Faster rates and adaptivity via learning rate separation
2022. 36th Conference on Neural Information Processing Systems (NeurIPS 2022), New Orleans, USA, November 28 - December 9, 2022.Adaptive Stochastic Variance Reduction for Non-convex Finite-Sum Minimization
2022. 36th Conference on Neural Information Processing Systems (NeurIPS), New Orleans, Louisianna, USA, November 28-December 9, 2022.Extra Newton: A First Approach to Noise-Adaptive Accelerated Second-Order Methods
2022. 36th Conference on Neural Information Processing Systems (NeurIPS), New Orleans, Louisianna, USA, November 28-December 9, 2022.Extrapolation and Spectral Bias of Neural Nets with Hadamard Product: a Polynomial Net Study
2022. 36th Conference on Neural Information Processing Systems (NeurIPS), New Orleans, USA, November 28 - December 3, 2022.Predicting in Uncertain Environments: Methods for Robust Machine Learning
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9118.Generalization Properties of NAS under Activation and Skip Connection Search
2022. 36th Conference on Neural Information Processing Systems - NeurIPS 2022, New Orleans, USA, November 28 - December 3, 2022.Robustness in deep learning: The good (width), the bad (depth), and the ugly (initialization)
2022. 36th Conference on Neural Information Processing Systems - NeurIPS 2022, New Orleans, USA, November 28 - December 3, 2022.On the Double Descent of Random Features Models Trained with SGD
2022. 36th Conference on Neural Information Processing Systems (NeurIPS 2022), New Orleans, USA, November 28 - December 3, 2022.Understanding Deep Neural Function Approximation in Reinforcement Learning via ϵ-Greedy Exploration
2022. Thirty-sixth Conference on Neural Information Processing Systems - NeurIPS 2022, New Orleans, USA, November 28 - December 3, 2022.Proximal Point Imitation Learning
2022. 36th Conference on Neural Information Processing Systems (NeurIPS), New Orleans, USA, November 28 - December 3, 2022.Identifiability and Generalizability from Multiple Experts in Inverse Reinforcement Learning
2022. 36th Conference on Neural Information Processing Systems (NeurIPS), New Orleans, USA, November 28 - December 3, 2022.Sound and Complete Verification of Polynomial Networks
2022. 36th Conference on Neural Information Processing Systems (NeurIPS), New Orleans, USA, November 28 - December 3, 2022.On The Convergence Of Stochastic Primal-Dual Hybrid Gradient
Siam Journal On Optimization. 2022-01-01. DOI : 10.1137/19M1296252.Score Matching Enables Causal Discovery of Nonlinear Additive Noise Models
2022. 38th International Conference on Machine Learning (ICML), Baltimore, Maryland, USA, July 17-23, 2022.Learning to sample in Cartesian MRI
Lausanne, EPFL, 2022. DOI : 10.5075/epfl-thesis-9981.A 16-Channel Neural Recording System-on-Chip With CHT Feature Extraction Processor in 65-nm CMOS
IEEE Journal of Solid-State Circuits. 2022. DOI : 10.1109/JSSC.2022.3161296.Faster One-Sample Stochastic Conditional Gradient Method for Composite Convex Minimization
2022. 25th International Conference on Artificial Intelligence and Statistics (AISTATS 2022), [ Virtual only] Valencia, Spain, March 28-30, 2022.High Probability Bounds for a Class of Nonconvex Algorithms with AdaGrad Stepsize
2022. 10th International Conference on Learning Representations (ICLR), Virtual, April 25-29, 2022.Controlling the Complexity and Lipschitz Constant improves Polynomial Nets
2022. 10th International Conference on Learning Representations (ICLR), Virtual, April 25-29, 2022.The spectral bias of polynomial neural networks
2022. 10th International Conference on Learning Representations (ICLR), Virtual, April 25-29, 2022.Escaping limit cycles: Global convergence for constrained nonconvex-nonconcave minimax problems
2022. 10th International Conference on Learning Representations (ICLR 2022), Virtual, April 25-29, 2022.A Newton Frank-Wolfe method for constrained self-concordant minimization
Journal Of Global Optimization. 2022. DOI : 10.1007/s10898-021-01105-z.Forward-reflected-backward method with variance reduction
Computational Optimization and Applications. 2021-08-19. DOI : 10.1007/s10589-021-00305-3.A 16-Channel Wireless Neural Recording System-on-Chip with CHT Feature Extraction Processor in 65nm CMOS
2021-05-17. 2021 IEEE Custom Integrated Circuits Conference (CICC), Virtual, April 25-30, 2021. DOI : 10.1109/CICC51472.2021.9431458.An Optimal-Storage Approach to Semidefinite Programming Using Approximate Complementarity
Siam Journal On Optimization. 2021-01-01. DOI : 10.1137/19M1244603.A Plug-and-Play Deep Image Prior
2021. International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2021), Toronto, Canada, June 6-11, 2021. DOI : 10.1109/ICASSP39728.2021.9414879.Sifting through the Noise: Universal First-Order Methods for Stochastic Variational Inequalities
2021. NeurIPS 2021 : Thirty-fifth Conference on Neural Information Processing Systems, Sydney, Australia [Virtual only], December 6-14, 2021.Convergence of adaptive algorithms for constrained weakly convex optimization
2021. 35th Conference on Neural Information Processing Systems (NeurIPS 2021), Sydney, Australia, December 6-14, 2021.STORM+: Fully Adaptive SGD with Momentum for Nonconvex Optimization
2021. 35th Conference on Neural Information Processing Systems (NeurIPS 2021), Sydney, Australia, December 6-14, 2021.A first-order primal-dual method with adaptivity to local smoothness
2021. 35th Conference on Neural Information Processing Systems (NeurIPS 2021), Sydney, Australia, December 6-14, 2021.Robust Inverse Reinforcement Learning under Transition Dynamics Mismatch
2021. 35th Conference on Neural Information Processing Systems (NeurIPS 2021), Sydney, Australia, December 6-14, 2021.The Effect of the Intrinsic Dimension on the Generalization of Quadratic Classifiers
2021. 35th Conference on Neural Information Processing Systems (NeurIPS 2021), Sydney, Australia, December 6-14, 2021.Subquadratic Overparameterization for Shallow Neural Networks
2021. 35th Conference on Neural Information Processing Systems (NeurIPS 2021), Sydney, Australia, December 6-14, 2021.Regret Minimization in Stochastic Non-Convex Learning via a Proximal-Gradient Approach
2021-01-01. International Conference on Machine Learning (ICML), ELECTR NETWORK, Jul 18-24, 2021.Asynchronous sar adc with unit length capacitors and constant common mode monotonic switching
WO2021161163 . 2021.Adaptation in Stochastic Algorithms: From Nonsmooth Optimization to Min-Max Problems and Beyond
Lausanne, EPFL, 2021. DOI : 10.5075/epfl-thesis-8120.Resource Trade-Offs in Circuits and Systems: from Neurotechnology to Communications
Lausanne, EPFL, 2021. DOI : 10.5075/epfl-thesis-9082.Kernel Conjugate Gradient Methods with Random Projections
Applied and Computational Harmonic Analysis. 2021. DOI : 10.1016/j.acha.2021.05.004.The Limits of Min-Max Optimization Algorithms: Convergence to Spurious Non-Critical Sets
2021. 38th International Conference on Machine Learning (ICML 2021), Online, July 18-24, 2021. p. 4337-4348.A reflected forward-backward splitting method for monotone inclusions involving Lipschitzian operators
Set-valued and Variational analysis. 2021. DOI : 10.1007/s11228-020-00542-4.Scalable Semidefinite Programming
SIAM Journal on Mathematics of Data Science. 2021. DOI : 10.1137/19M1305045.An Introductory Guide to Fano's Inequality with Applications in Statistical Estimation
Information-Theoretic Methods in Data Science; Cambridge University Press, 2021. p. 487-528.Robust Reinforcement Learning via Adversarial training with Langevin Dynamics
2020-11-05An AC-Coupled Wideband Neural Recording Front-End With Sub-1 mm2×fJ/conv-step Efficiency and 0.97 NEF
IEEE Solid-State Circuits Letters. 2020-08-04. DOI : 10.1109/LSSC.2020.3013993.Machine Learning From Distributed, Streaming Data [From the Guest Editors]
Ieee Signal Processing Magazine. 2020-05-01. DOI : 10.1109/MSP.2020.2972654.Optimization for Reinforcement Learning: From a single agent to cooperative agents
Ieee Signal Processing Magazine. 2020-05-01. DOI : 10.1109/MSP.2020.2976000.Lipschitz constant estimation for Neural Networks via sparse polynomial optimization
2020-04-26. 8th International Conference on Learning Representations, Addis Ababa, ETHIOPIA, April 26-30, 2020.On the Almost Sure Convergence of Stochastic Gradient Descent in Non-Convex Problems
2020. 34th Conference on Neural Information Processing Systems (NeurIPS 2020), Virtual, December 6-12, 2020.Convergence without Convexity: Sampling, Optimization, and Games
Lausanne, EPFL, 2020. DOI : 10.5075/epfl-thesis-10349.Optimal Convergence for Distributed Learning with Stochastic Gradient Methods and Spectral Algorithms
Journal of Machine Learning Research. 2020.Random extrapolation for primal-dual coordinate descent
2020. 37th International Conference on Machine Learning (ICML 2020), Online, July 13-18, 2020.Conditional gradient methods for stochastically constrained convex minimization
2020. 37th International Conference on Machine Learning (ICML), virtual, July 12-18, 2020.Efficient Proximal Mapping of the 1-path-norm of Shallow Networks
2020. 37th International Conference on Machine Learning (ICML), Virtual, July 13-18, 2020.A new regret analysis for Adam-type algorithms
2020. 37th International Conference on Machine Learning (ICLM 2020), Virtual, July 13-18, 2020.Double-Loop Unadjusted Langevin Algorithm
2020. 37th International Conference on Machine Learning (ICLM 2020), Virtual, July 12-18, 2020.Wasserstein Distributionally Robust Learning
Lausanne, EPFL, 2020. DOI : 10.5075/epfl-thesis-10012.Scalable Learning-Based Sampling Optimization For Compressive Dynamic MRI
2020. International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Barcelona, Spain, May 4-8, 2020. p. 8584-8588. DOI : 10.1109/ICASSP40776.2020.9053345.Convergences of Regularized Algorithms and Stochastic Gradient Methods with Random Projections
Journal of Machine Learning Research. 2020.Personalizable intervention systems to promote healthy behavior change
Lausanne, EPFL, 2020. DOI : 10.5075/epfl-thesis-10008.An adaptive primal-dual framework for nonsmooth convex minimization
Mathematical Programming Computation. 2020. DOI : 10.1007/s12532-019-00173-3.Optimal Rates for Spectral Algorithms with Least-Squares Regression over Hilbert Spaces
Applied and Computational Harmonic Analysis. 2020. DOI : 10.1016/j.acha.2018.09.009.Interactive Teaching Algorithms for Inverse Reinforcement Learning
2019-08-10. The 28th International Joint Conference on Artificial Intelligence, 2019., Macao, China, August 10-16, 2019.Interactive Teaching Algorithms for Inverse Reinforcement Learning
2019-08-10. 28th International Joint Conference on Artificial Intelligence, 2019, Macao, China, August 10-16, 2019. p. 2692–2700.Chemical machine learning with kernels: The impact of loss functions
International Journal Of Quantum Chemistry. 2019-05-05. DOI : 10.1002/qua.25872.Inertial Three-Operator Splitting Method and Applications
SIAM Conference on Optimization - OP17, Vancouver, British Columbia, Canada, May 22-25, 2017.Ultrasensitive hyperspectral imaging and biodetection enabled by dielectric metasurfaces
Nature Photonics. 2019-04-01. DOI : 10.1038/s41566-019-0394-6.On the convergence of stochastic primal-dual hybrid gradient
2019Data-driven Measurement Designs for Magnetic Resonance Imaging
Lausanne, EPFL, 2019. DOI : 10.5075/epfl-thesis-9651.Streaming Low-Rank Matrix Approximation With An Application To Scientific Simulation
SIAM Journal on Scientific Computing. 2019-01-01. DOI : 10.1137/18M1201068.Stochastic Frank-Wolfe for Composite Convex Minimization
2019. NeurIPS 2019 : Thirty-third Conference on Neural Information Processing Systems, Vancouver, Canada, December 8-14, 2019.An Inexact Augmented Lagrangian Framework for Nonconvex Optimization with Nonlinear Constraints
2019. NeurIPS 2019 : Thirty-third Conference on Neural Information Processing Systems, Vancouver, Canada, December 8-14, 2019.Fast and Provable ADMM for Learning with Generative Priors
2019. 33rd Conference on Neural Information Processing Systems (NeurIPS), Vancouver, Canada, December 8-14, 2019.UniXGrad: A Universal, Adaptive Algorithm with Optimal Guarantees for Constrained Optimization
2019. 33rd Conference on Neural Information Processing Systems (NeurIPS), Vancouver, Canada, December 8-14, 2019.Scalable Convex Optimization Methods for Semidefinite Programming
Lausanne, EPFL, 2019. DOI : 10.5075/epfl-thesis-9598.On Certifying Non-Uniform Bounds against Adversarial Attacks
2019. 36th International Conference on Machine Learning (ICML)'2019, Long Beach, USA, June 9-15, 2019.Conditional Gradient Methods via Stochastic Path-Integrated Differential Estimator
2019. 36th International Conference on Machine Learning (ICML 2019), Long Beach, USA, June 9-15, 2019.Efficient learning of smooth probability functions from Bernoulli tests with guarantees.
2019. 36th International Conference on Machine Learning (ICML 2019), Long Beach, USA, June 9-15, 2019.A Conditional Gradient-Based Augmented Lagrangian Framework
2019. 36th International Conference on Machine Learning (ICML 2019), Long Beach, USA, June 9-15, 2019.Almost surely constrained convex optimization
2019. 36th International Conference on Machine Learning (ICML 2019), Long Beach, USA, June 9-15, 2019.Teaching & PhD
Teaching
Electrical and Electronics Engineering
PhD Students
Abad Rocamora Elias, Abranches Figueiredo Simoes De Carvalho Pedro, Afzal Arshia, Dadi Leello Tadesse, Janchevski Andrej, Krawczuk Igor, Pethick Thomas Michaelsen, Viano Luca, Wu Yongtao, Xie Wanyun, Zhu Zhenyu,Past EPFL PhD Students
Alacaoglu Ahmet , Aprile Cosimo , Bogunovic Ilija , El Halabi Marwa , Gözcü Baran , Hsieh Ya-Ping , Karimi Mahabadi Rabeeh , Kavis Ali , Kyrillidis Anastasios , Latorre Gomez Fabian Ricardo , Li Yen-Huan , Rolland Paul Thierry Yves , Sahin Mehmet Fatih , Sanchez Thomas , Uran Arda , Yurtsever Alp ,Courses
Mathematics of data: from theory to computation
This course provides an overview of key advances in continuous optimization and statistical analysis for machine learning. We review recent learning formulations and models as well as their guarantees, describe scalable solution techniques and algorithms, and illustrate the trade-offs involved.
Reinforcement learning
This course describes theory and methods for Reinforcement Learning (RL), which revolves around decision making under uncertainty. The course covers classic algorithms in RL as well as recent algorithms under the lens of contemporary optimization.
EECS Seminar: Advanced Topics in Machine Learning
Students learn about advanced topics in machine learning, artificial intelligence, optimization, and data science. Students also learn to interact with scientific work, analyze and understand strengths and weaknesses of scientific arguments of both theoretical and experimental results.