Konstantin Mishchenko

King Abdullah University of Science and Technology

H-index: 20

Asia-Saudi Arabia

About Konstantin Mishchenko

Konstantin Mishchenko, With an exceptional h-index of 20 and a recent h-index of 20 (since 2020), a distinguished researcher at King Abdullah University of Science and Technology, specializes in the field of Deep Learning, Optimization, Federated Learning.

His recent articles reflect a diverse array of research interests and contributions to the field:

Super-universal regularized newton method

Two Losses Are Better Than One: Faster Optimization Using a Cheaper Proxy

Regularized Newton Method with Global Convergence

Learning-Rate-Free Learning by D-Adaptation

Convergence of first-order algorithms for meta-learning with Moreau envelopes

Server-side stepsizes and sampling without replacement provably help in federated optimization

Prodigy: An expeditiously adaptive parameter-free learner

Federated Learning Under Second-Order Data Heterogeneity

Konstantin Mishchenko Information

University	King Abdullah University of Science and Technology
Position	___
Citations(all)	1746
Citations(since 2020)	1738
Cited By	395
hIndex(all)	20
hIndex(since 2020)	20
i10Index(all)	26
i10Index(since 2020)	26
Email	Access Email
University Profile Page	King Abdullah University of Science and Technology
Google Scholar	View Google Scholar Profile

Konstantin Mishchenko Skills & Research Interests

Deep Learning

Optimization

Federated Learning

Top articles of Konstantin Mishchenko

Title	Journal	Author(s)	Publication Date
Super-universal regularized newton method	SIAM Journal on Optimization	Nikita Doikov Konstantin Mishchenko Yurii Nesterov	2024/3/31
Two Losses Are Better Than One: Faster Optimization Using a Cheaper Proxy		Blake Woodworth Konstantin Mishchenko Francis Bach	2023/7
Regularized Newton Method with Global Convergence	SIAM Journal on Optimization	Konstantin Mishchenko	2023/3
Learning-Rate-Free Learning by D-Adaptation		Aaron Defazio Konstantin Mishchenko	2023/7
Convergence of first-order algorithms for meta-learning with Moreau envelopes	arXiv preprint arXiv:2301.06806	Konstantin Mishchenko Slavomir Hanzely Peter Richtárik	2023/1/17
Server-side stepsizes and sampling without replacement provably help in federated optimization		Grigory Malinovsky Konstantin Mishchenko Peter Richtárik	2023/12/8
Prodigy: An expeditiously adaptive parameter-free learner	arXiv preprint arXiv:2306.06101	Konstantin Mishchenko Aaron Defazio	2023/6/9
Federated Learning Under Second-Order Data Heterogeneity		Konstantin Mishchenko Rui Li Hongxiang Fan Stylianos Venieris	2023/10/13
When, Why and How Much? Adaptive Learning Rate Scheduling by Refinement	arXiv preprint arXiv:2310.07831	Aaron Defazio Ashok Cutkosky Harsh Mehta Konstantin Mishchenko	2023/10/11
Partially personalized federated learning: Breaking the curse of data heterogeneity	arXiv preprint arXiv:2305.18285	Konstantin Mishchenko Rustem Islamov Eduard Gorbunov Samuel Horváth	2023/5/29
Adaptive proximal gradient method for convex optimization	arXiv preprint arXiv:2308.02261	Yura Malitsky Konstantin Mishchenko	2023/8/4
DoWG Unleashed: An Efficient Universal Parameter-Free Gradient Descent Method	NeurIPS 2023	Ahmed Khaled Konstantin Mishchenko Chi Jin	2023/5/25
Proximal and Federated Random Reshuffling		Konstantin Mishchenko Ahmed Khaled Peter Richtárik	2022/6/28
ProxSkip: Yes! Local Gradient Steps Provably Lead to Communication Acceleration! Finally!		Konstantin Mishchenko Grigory Malinovsky Sebastian Stich Peter Richtárik	2022/6/28
Asynchronous SGD Beats Minibatch SGD under Arbitrary Delays		Konstantin Mishchenko Francis Bach Mathieu Even Blake Woodworth	2022/12
IntSGD: Adaptive Floatless Compression of Stochastic Gradients	ICLR 2022 - International Conference on Learning Representations	Konstantin Mishchenko Bokun Wang Dmitry Kovalev Peter Richtárik	2022
Dualize, split, randomize: Toward fast nonsmooth optimization algorithms	Journal of Optimization Theory and Applications	Adil Salim Laurent Condat Konstantin Mishchenko Peter Richtárik	2022/10
Stochastic distributed learning with gradient quantization and double-variance reduction	Optimization Methods and Software	Samuel Horváth Dmitry Kovalev Konstantin Mishchenko Peter Richtárik Sebastian Stich	2023/1/2
Adaptive learning rates for faster stochastic gradient methods	arXiv preprint arXiv:2208.05287	Samuel Horváth Konstantin Mishchenko Peter Richtárik	2022/8/10
Adaptive gradient descent without descent		Yura Malitsky Konstantin Mishchenko	2020