Avetik G. Karagulyan

According to our database, Avetik G. Karagulyan authored at least 8 papers between 2017 and 2024.

Collaborative distances:

• Dijkstra number of five.
• Erdős number of four.

2024

SPAM: Stochastic Proximal Point Method with Momentum Variance Reduction for Non-convex Cross-Device Federated Learning.

Avetik G. Karagulyan, Egor Shulgin, Abdurakhmon Sadiev, Peter Richtárik

CoRR, 2024

Langevin Monte Carlo for strongly log-concave distributions: Randomized midpoint revisited.

Lu Yu, Avetik G. Karagulyan, Arnak S. Dalalyan

Proceedings of the Twelfth International Conference on Learning Representations, 2024

Det-CGD: Compressed Gradient Descent with Matrix Stepsizes for Non-Convex Optimization.

Hanmin Li, Avetik G. Karagulyan, Peter Richtárik

Proceedings of the Twelfth International Conference on Learning Representations, 2024

2023

ELF: Federated Langevin Algorithms with Primal, Dual and Bidirectional Compression.

Avetik G. Karagulyan, Peter Richtárik

CoRR, 2023

Convergence of Stein Variational Gradient Descent under a Weaker Smoothness Condition.

Lukang Sun, Avetik G. Karagulyan, Peter Richtárik

Proceedings of the International Conference on Artificial Intelligence and Statistics, 2023

2022

Bounding the Error of Discretized Langevin Algorithms for Non-Strongly Log-Concave Targets.

Arnak S. Dalalyan, Avetik G. Karagulyan, Lionel Riou-Durand

J. Mach. Learn. Res., 2022

2020

Penalized Langevin dynamics with vanishing penalty for smooth and log-concave targets.

Avetik G. Karagulyan, Arnak S. Dalalyan

Proceedings of the Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, 2020

2017

User-friendly guarantees for the Langevin Monte Carlo with inaccurate gradient.

Arnak S. Dalalyan, Avetik G. Karagulyan

CoRR, 2017