Ali Kavis

27 papers A* 15Journal 12
YearRankTypeTitle / Venue / Authors
2025 J jnl
Online Learning-guided Learning Rate Adaptation via Gradient Alignment.
CoRR
Ruichen Jiang, Ali Kavis, Aryan Mokhtari
2025 A* conf
Upweighting Easy Samples in Fine-Tuning Mitigates Forgetting.
ICML
Sunny Sanyal, Hayden Prairie, Rudrajit Das, Ali Kavis, Sujay Sanghavi
2025 J jnl
Upweighting Easy Samples in Fine-Tuning Mitigates Forgetting.
CoRR
Sunny Sanyal, Hayden Prairie, Rudrajit Das, Ali Kavis, Sujay Sanghavi
2024 A* conf
Adaptive and Optimal Second-order Optimistic Methods for Minimax Optimization.
NeurIPS
Ruichen Jiang, Ali Kavis, Qiujiang Jin, Sujay Sanghavi, Aryan Mokhtari
2024 J jnl
Adaptive and Optimal Second-order Optimistic Methods for Minimax Optimization.
CoRR
Ruichen Jiang, Ali Kavis, Qiujiang Jin, Sujay Sanghavi, Aryan Mokhtari
2024 A* conf
Advancing the Lower Bounds: an Accelerated, Stochastic, Second-order Method with Optimal Adaptation to Inexactness.
ICLR
Artem Agafonov, Dmitry Kamzolov, Alexander V. Gasnikov, Ali Kavis, Kimon Antonakopoulos, Volkan Cevher, Martin Takác
2024 J jnl
Understanding Contrastive Learning via Gaussian Mixture Models.
CoRR
Parikshit Bansal, Ali Kavis, Sujay Sanghavi
2024 A* conf
Universal Gradient Methods for Stochastic Convex Optimization.
ICML
Anton Rodomanov, Ali Kavis, Yongtao Wu, Kimon Antonakopoulos, Volkan Cevher
2023 A* conf
Alternation makes the adversary weaker in two-player games.
NeurIPS
Volkan Cevher, Ashok Cutkosky, Ali Kavis, Georgios Piliouras, Stratis Skoulakis, Luca Viano
2022 A* conf
Adaptive Stochastic Variance Reduction for Non-convex Finite-Sum Minimization.
NeurIPS
Ali Kavis, Stratis Skoulakis, Kimon Antonakopoulos, Leello Tadesse Dadi, Volkan Cevher
2022 J jnl
Adaptive Stochastic Variance Reduction for Non-convex Finite-Sum Minimization.
CoRR
Ali Kavis, Stratis Skoulakis, Kimon Antonakopoulos, Leello Tadesse Dadi, Volkan Cevher
2022 A* conf
Extra-Newton: A First Approach to Noise-Adaptive Accelerated Second-Order Methods.
NeurIPS
Kimon Antonakopoulos, Ali Kavis, Volkan Cevher
2022 J jnl
Extra-Newton: A First Approach to Noise-Adaptive Accelerated Second-Order Methods.
CoRR
Kimon Antonakopoulos, Ali Kavis, Volkan Cevher
2022 A* conf
High Probability Bounds for a Class of Nonconvex Algorithms with AdaGrad Stepsize.
ICLR
Ali Kavis, Kfir Yehuda Levy, Volkan Cevher
2022 J jnl
High Probability Bounds for a Class of Nonconvex Algorithms with AdaGrad Stepsize.
CoRR
Ali Kavis, Kfir Yehuda Levy, Volkan Cevher
2021 J jnl
STORM+: Fully Adaptive SGD with Momentum for Nonconvex Optimization.
CoRR
Kfir Y. Levy, Ali Kavis, Volkan Cevher
2021 A* conf
STORM+: Fully Adaptive SGD with Recursive Momentum for Nonconvex Optimization.
NeurIPS
Kfir Y. Levy, Ali Kavis, Volkan Cevher
2021 A* conf
Sifting through the noise: Universal first-order methods for stochastic variational inequalities.
NeurIPS
Kimon Antonakopoulos, Thomas Pethick, Ali Kavis, Panayotis Mertikopoulos, Volkan Cevher
2020 A* conf
Double-Loop Unadjusted Langevin Algorithm.
ICML
Paul Rolland, Armin Eftekhari, Ali Kavis, Volkan Cevher
2020 J jnl
Double-Loop Unadjusted Langevin Algorithm.
CoRR
Paul Rolland, Armin Eftekhari, Ali Kavis, Volkan Cevher
2020 A* conf
On the Almost Sure Convergence of Stochastic Gradient Descent in Non-Convex Problems.
NeurIPS
Panayotis Mertikopoulos, Nadav Hallak, Ali Kavis, Volkan Cevher
2020 J jnl
On the Almost Sure Convergence of Stochastic Gradient Descent in Non-Convex Problems.
CoRR
Panayotis Mertikopoulos, Nadav Hallak, Ali Kavis, Volkan Cevher
2019 A* conf
Efficient learning of smooth probability functions from Bernoulli tests with guarantees.
ICML
Paul Rolland, Ali Kavis, Alexander Immer, Adish Singla, Volkan Cevher
2019 A* conf
UniXGrad: A Universal, Adaptive Algorithm with Optimal Guarantees for Constrained Optimization.
NeurIPS
Ali Kavis, Kfir Y. Levy, Francis R. Bach, Volkan Cevher
2019 J jnl
UniXGrad: A Universal, Adaptive Algorithm with Optimal Guarantees for Constrained Optimization.
CoRR
Ali Kavis, Kfir Y. Levy, Francis R. Bach, Volkan Cevher
2018 J jnl
Efficient learning of smooth probability functions from Bernoulli tests with guarantees.
CoRR
Paul Rolland, Ali Kavis, Adish Singla, Volkan Cevher
2018 A* conf
Mirrored Langevin Dynamics.
NeurIPS
Ya-Ping Hsieh, Ali Kavis, Paul Rolland, Volkan Cevher