Dmitrii Avdiukhin


Dmitrii Avdiukhin

I am a McCormick Postdoctoral Fellow at Northwestern University. Mentor: Konstantin Makarychev

I got my Ph.D. at Indiana University in 2023. Advisor: Grigory Yaroslavtsev

E-mail: first name (at) first name (dot) us

Looking for a Faculty/Postdoctoral/Research Scientist Position starting Fall 2025

My CV (pdf)


Research interests

  • Convex and nonconvex optimization
  • Theoretical foundations of machine learning
  • Hierarchical clustering
  • Approximation algorithms
  • Learning Theory

News

26 Oct 2024
I am one of the organizers of Junior Theorists Workshop 2024, held jointly by Northwestern University (December 5) and TTIC (December 6).
10 Oct 2024
Our paper "Embedding Dimension of Contrastive Learning and k-Nearest Neighbors" is accepted for NeurIPS 24!
16 Jan 2024
Our paper "Optimal Sample Complexity of Contrastive Learning" is accepted for ICLR 24 for spotlight presentation!
8 Dec 2023
Our paper "Approximation Scheme for Weighted Metric Clustering via Sherali-Adams" is accepted for AAAI 24!
30 Nov 2023
Junior Theorists Workshop 2023 starts today! The first day is held at Northwestern University, and the second day is held by TTIC. We will be hosting some of the best PhD students and postdocs, so look forward to excellent talks!
8 Sep 2023
I'm excited to join Northwestern University as a McCormick Postdoctoral Fellow under the mentorship of Konstantin Makarychev!

Selected Publications

[Check this link for the list of all publications]
ICLR 2024
N. Alon, D. Avdiukhin, D. Elboim, O. Fischer, G. Yaroslavtsev. "Optimal Sample Complexity of Contrastive Learning"
AAAI 2023
D. Avdiukhin, G. Yaroslavtsev, D. Vainstein, O. Fischer, S. Das, and F. Mirza. "Tree Learning: Optimal Algorithms and Sample Complexity" [paper]
NeurIPS 2021
D. Avdiukhin., and G. Yaroslavtsev. "Escaping Saddle Points with Compressed SGD" [paper]
ICML 2021
D. Avdiukhin., and S. Kasiviswanathan. "Federated Learning under Arbitrary Communication Patterns" [paper]
AAAI 2021
D. Avdiukhin., S. Naumov, and G. Yaroslavtsev. "Objective-Based Hierarchical Clustering of Deep Embedding Vectors" [paper]
VLDB 2019
D. Avdiukhin, S. Pupyrev and G. Yaroslavtsev. “Multi-Dimensional Balanced Graph Partitioning via Projected Gradient Descent” [paper]

Experience

Summer 2022
Research Intern, Amazon.
Demonstration selection for few-shot learning for small language models.
Summer 2020
Research Intern, Amazon.
Federated Learning under weak assumptions
Summer 2019
Research Intern, Amazon, New York.
Improving accuracy and performance of graph convolutional networks
Summer 2018
Software Engineer, Pro Unlimited @ Facebook, Menlo Park.
Working on balanced graph partitioning
2016-2017
Researcher, ITMO University.
Model generation from execution traces
2013-2016
Software Engineer. JetBrains, Saint Petersburg.
SQL dialects support
2012-2013
Software Engineer. Lanit Tercom, Saint Petersburg.
Participating in project of migration a system from SQL server to Oracle

Organizer

November 2023
Junior Theorists Workshop 2023
December 2024
Junior Theorists Workshop 2024

Talks and Posters

ICLR 2024
Poster
“Optimal Sample Complexity of Contrastive Learning”
ITA 2023
Talk
“First-Order Methods in Distributed Optimization”
OPT 2022
Poster
“HOUDINI: Escaping from Moderately Constrained Saddles”
OPT 2022
Poster
“Bidirectional Adaptive Communication for Heterogeneous Distributed Learning”
NeurIPS 2021
Poster
“Escaping Saddle Points with Compressed SGD”
OPT 2020
Paster
“Escaping Saddle Points with Compressed SGD”
VLDB 2019
Talk
“Multi-Dimensional Balanced Graph Partitioning via Projected Gradient Descent”
KDD 2019
Talk
“Adversarially Robust Submodular Maximization under Knapsack Constraints”

Other Talks

Junior Theorists Workshop 2023
"Optimal Sample Complexity of Contrastive Learning"
SIAM OP 2023
"Escaping Saddle Points with Compressed SGD"
Google Algorithms Seminar
"Tree Learning: Optimal Algorithms and Sample Complexity"
SPbSU, Russia
"Escaping from Saddle Points with Compressed SGD"
Yandex, Russia
"Multi-Dimensional Balanced Graph Partitioning via Projected Gradient Descent"

Fellowships

2019
Nominated for Google PhD Fellowship Program by Indiana University
2019
Nominated for Microsoft Fellowship by Indiana University

Teaching

I'm currently teaching Design & Analysis of Algorithms at Northwestern University
Check this link for my teaching experience.