#31
Talks and posters
Invited talks, conference talks, and seminars
#30
#29
High-Probability Convergence for Composite and Distributed Stochastic Minimization and Variational Inequalities with Heavy-Tailed Noise
#28
Byzantine Robustness and Partial Participation Can Be Achieved Simultaneously: Just Clip Gradient Differences
#27
Last-Iterate Convergence of Extragradient-Based Methods
#26
Byzantine Robustness and Partial Participation Can Be Achieved Simultaneously: Just Clip Gradient Differences
#25
Byzantine Robustness and Partial Participation Can Be Achieved Simultaneously: Just Clip Gradient Differences
#24
Byzantine Robustness and Partial Participation Can Be Achieved Simultaneously: Just Clip Gradient Differences
(Keynote talk)
#23
#22
Clipped Methods for Stochastic Optimization with Heavy-Tailed Noise
#21
Algorithms for Stochastic Optimization with Heavy-Tailed Noise and Connections with the Training of Large Language Models
#20
Earlier talks and seminars19 archived talks, 2017-2022
#19
Variance Reduction is an Antidote to Byzantines: Better Rates, Weaker Assumptions and Communication Compression as a Cherry on the Top
#18
#17
Distributed Methods with Absolute Compression and Error Compensation
#16
Secure Distributed Training at Scale
#15
#14
Moshpit SGD: Communication-Efficient Decentralized Training on Heterogeneous Unreliable Devices
#13
Moshpit SGD: Communication-Efficient Decentralized Training on Heterogeneous Unreliable Devices
#12
Extragradient Method: $O(1/K)$ Last-Iterate Convergence for Monotone Variational Inequalities and Connections with Cocoercivity
#11
#10
#9
#8
#7
#6
#5
A Unified Theory of SGD: Variance Reduction, Sampling, Quantization and Coordinate Descent
#4
23rd International Symposium on Mathematical Programming
Talk An Accelerated Directional Derivative Method for Smooth Stochastic Convex Optimization
Bordeaux, 6 July, 2018
#3
#2
60th Scientific Conference of MIPT
Talk About accelerated Directional Search with non-Euclidean prox-structure
Moscow, Russia, 25 November, 2017
#1
Conference posters and presentation materials
#34
Methods for Convex $(L_0,L_1)$-Smooth Optimization: Clipping, Acceleration, and Adaptivity
Singapore, 26 April, 2025
#33
Methods with Local Steps and Random Reshuffling for Generally Smooth Non-Convex Federated Optimization
Singapore, 25 April, 2025
#32
Byzantine Robustness and Partial Participation Can Be Achieved at Once: Just Clip Gradient Differences
New Orleans, USA, 11 December, 2024
#31
Federated Optimization Algorithms with Random Reshuffling and Gradient Compression
New Orleans, USA, 13 December, 2024
#30
#29
Remove that Square Root: A New Efficient Scale-Invariant Version of AdaGrad
New Orleans, USA, 11 December, 2024
#28
#27
Communication Compression for Byzantine Robust Learning: New Efficient Algorithms and Improved Rates
Valencia, Spain, 3 May, 2024
#26
Breaking the Heavy-Tailed Noise Barrier in Stochastic Optimization Problems
Valencia, Spain, 3 May, 2024
#25
Single-Call Stochastic Extragradient Methods for Structured Non-monotone Variational Inequalities: Improved Analysis under Weaker Conditions
New Orleans, USA, 10 December - 16 December, 2023
#24
Accelerated Zeroth-order Method for Non-Smooth Stochastic Convex Optimization Problem with Infinite Variance
New Orleans, USA, 10 December - 16 December, 2023
#23
Byzantine-Tolerant Methods for Distributed Variational Inequalities
New Orleans, USA, 10 December - 16 December, 2023
#22
High-Probability Bounds for Stochastic Optimization and Variational Inequalities: the Case of Unbounded Variance
Honolulu, USA, 27 July, 2023
#21
#20
#19
Stochastic Gradient Descent-Ascent: Unified Theory and New Efficient Methods
Valencia, Spain, 27 April, 2023
Earlier poster presentations18 archived posters, 2018-2022
#18
Clipped Stochastic Methods for Variational Inequalities with Heavy-Tailed Noise
New Orleans, USA, 28 November - 9 December, 2022
#17
Last-Iterate Convergence of Optimistic Gradient Method for Monotone Variational Inequalities
New Orleans, USA, 28 November - 9 December, 2022
#16
#15
3PC: Three Point Compressors for Communication-Efficient Distributed Training and a Better Theory for Lazy Aggregation
Baltimore, USA, 21 July, 2022
#14
#13
#12
Moshpit SGD: Communication-Efficient Decentralized Training on Heterogeneous Unreliable Devices
Online, 10 December, 2021
#11
#10
#9
Stochastic Optimization with Heavy-Tailed Noise via Accelerated Gradient Clipping
Online, 6-12 December, 2020
#8
#7
A Unified Theory of SGD: Variance Reduction, Sampling, Quantization and Coordinate Descent
Online, 26-28 August, 2020
#6
#5
#4
A Stochastic Derivative Free Optimization Method with Momentum
Based on the joint work with Adel Bibi, Ozan Sener, El Houcine Bergou and Peter Richtárik
Vancouver, Canada, 14 December, 2019
#3
An Accelerated Method for Derivative-Free Smooth Stochastic Convex Optimization
Based on the joint work with Pavel Dvurechensky and Alexander Gasnikov
Vancouver, Canada, 13 December, 2019
#2
An Accelerated Directional Derivative Method for Smooth Stochastic Convex Optimization
Voronovo, Russia, 10-15 June, 2018
Selected for an accompanying talk and awarded third prize in the participant talk competition.
#1
Stochastic Spectral Descent Methods
Dmitry Kovalev, Eduard Gorbunov, Elnur Gasanov, Peter Richtárik
KAUST, Thuwal, KSA, 5 - 7 February, 2018