Volkan Cevher

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. He was also a Faculty Fellow in the Electrical and Computer Engineering Department at Rice University from 2010-2020. Currently, he is an Associate Professor at the Swiss Federal Institute of Technology Lausanne and an Amazon Scholar. His research interests include machine learning, optimization theory and methods, and automated control. Dr. Cevher is an IEEE Fellow ('24), an ELLIS fellow, and was the recipient of the ICML AdvML Best Paper Award in 2023, 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

Training Deep Learning Models with Norm-Constrained LMOs

T. Pethick; W. Xie; K. Antonakopoulos; Z. Zhu; A. Silveti-Falls et al.

2025. Forty-Second International Conference on Machine Learning, Vancouver, Canada, 2025-07-13 - 2025-07-19.

Volkan Cevher

Associate Professor

Fields of expertise

Biography

Publications

Infoscience publications

Selected Publications

Training Deep Learning Models with Norm-Constrained LMOs

Continuous-Time Analysis of Heavy Ball Momentum in Min-Max Games

Best of Both Worlds: Regret Minimization versus Minimax Play

Layer-wise Quantization for Quantized Optimistic Dual Averaging

Accelerating Spectral Clustering under Fairness Constraints

IL-SOAR : Imitation Learning with Soft Optimistic Actor cRitic

Generalization of Noisy SGD in Unbounded Non-convex Settings

CHAMELEON: A Flexible Data-mixing Framework for Language Model Pretraining and Finetuning

Adaptive Bilevel Optimization

Addressing Label Shift In Distributed Learning Via Entropy Regularization

Adversarial Training For Defense Against Label Poisoning Attacks

Quantum-Peft: Ultra Parameter-Efficient Fine-Tuning

How Gradient Descent Balances Features: A Dynamical Analysis For Two-Layer Neural Networks

Certified Robustness Under Bounded Levenshtein Distance

Efficient Interpolation Between Extragradient And Proximal Methods For Weak MVIS

Faster Inference Of Flow-Based Generative Models Via Improved Data-Noise Coupling

Single-pass Detection of Jailbreaking Input in Large Language Models

Hadamard product in deep learning: Introduction, Advances and Challenges

SAMPa: Sharpness-aware Minimization Parallelized

Membership Inference Attacks against Large Vision-Language Models

REST: Efficient and Accelerated EEG Seizure Analysis through Residual State Updates

Going beyond Compositions, DDPMs Can Produce Zero-Shot Interpolations

Learning with Norm Constrained, Over-parameterized, Two-layer Neural Networks

Truly No-Regret Learning in Constrained MDPs

Mixed Nash for Robust Federated Learning

Universal Gradient Methods for Stochastic Convex Optimization

Efficient local linearity regularization to overcome catastrophic overfitting

Multilinear Operator Networks

Generalization of Scaled Deep ResNets in the Mean-Field Regime

Imitation Learning in Discounted Linear MDPs without exploration assumptions

Efficient Continual Finite-Sum Minimization

Learning to Remove Cuts in Integer Linear Programming

On the Generalization of Stochastic Gradient Descent with Momentum

Improving SAM Requires Rethinking its Optimization Formulation

Revisiting Character-level Adversarial Attacks for Language Models

Robust NAS under adversarial training: benchmark, theory, and beyond

High-Dimensional Kernel Methods under Covariate Shift: Data-Dependent Implicit Regularization

Graph generative deep learning models with an application to circuit topologies

Stable Nonconvex-Nonconcave Training via Linear Interpolation

A unified stochastic approximation framework for learning in games

Improving Generalization of Pretrained Language Models

Maximum Independent Set: Self-Training through Dynamic Programming

Distributed Extra-Gradient With Optimal Complexity And Communication Guarantees

Efficient Online Clustering with Moving Costs

DiGress: Discrete Denoising diffusion for graph generation

Regularization of polynomial networks for image recognition

Universal and adaptive methods for robust stochastic optimization

Initialization Matters: Privacy-Utility Analysis of Overparameterized Neural Networks

Benign Overfitting in Deep Neural Networks under Lazy Training

Augmented Lagrangian Methods for Provable and Scalable Machine Learning

On the Convergence of Encoder-only Shallow Transformers

Semi Bandit Dynamics in Congestion Games: Convergence to Nash Equilibrium and No-Regret Guarantees.

What can online reinforcement learning with function approximation benefitfrom general coverage conditions

Finding Actual Descent Directions For Adversarial Training

Regularization of polynomial networks for image recognition

Alternation makes the adversary weaker in two-player games

Revisiting adversarial training for the worst-performing class

Adversarial Training Should Be Cast As a Non-Zero-Sum Game

When do Minimax-fair Learning and Empirical Risk Minimization Coincide?

Federated Learning under Covariate Shifts with Generalization Guarantees

Robust Training and Verification of Deep Neural Networks

Solving stochastic weak Minty variational inequalities without increasing batch size

Sample Complexity Bounds for Score-Matching: Causal Discovery and Generative Modeling

UNDERGRAD: A Universal Black-Box Optimization Method with Almost Dimension-Free Convergence Rate Guarantees

Predicting in Uncertain Environments: Methods for Robust Machine Learning

Sound and Complete Verification of Polynomial Networks

A Natural Actor-Critic Framework for Zero-Sum Markov Games

Extrapolation and Spectral Bias of Neural Nets with Hadamard Product: a Polynomial Net Study

A Newton Frank-Wolfe method for constrained self-concordant minimization

Robustness in deep learning: The good (width), the bad (depth), and the ugly (initialization)

Score Matching Enables Causal Discovery of Nonlinear Additive Noise Models

Identifiability and Generalizability from Multiple Experts in Inverse Reinforcement Learning

Adaptive Stochastic Variance Reduction for Non-convex Finite-Sum Minimization