Posts

Our session on Optimization for Deep Learning at MDS Conference

Peter Richtárik, Filip Hanzely and I are organizing a session on Optimization for Deep Learning at the SIAM Conference on Mathematics of Datascience (MDS20). See the link below for more info: https://www.siam.org/conferences/cm/conference/mds20

Our presenters are:

Simon Shaolei Du from the Institute for Advanced Study of Princeton (http://simonshaoleidu.com/)

Haihao Lu from Google Research NYC (http://web.mit.edu/haihao/www/)

Sai Praneeth Karimireddy from EPFL (https://people.epfl.ch/sai.karimireddy)

Filip Hanzely from KAUST (https://fhanzely.github.io/)

Last updated on Feb 13, 2022 1 min read

Adaptive Gradient got accepted to ICML

Our work with Yura Malitsky “Adaptive Gradient Descent without Descent” got accepted at ICML with scores ‘Accept’, ‘Accept’ and ‘Weak Accept’. You can see the arxiv version of the paper here: https://arxiv.org/abs/1910.09529. Also wait for our follow-up work that we hope to release in the next few months!

Last updated on Feb 13, 2022 1 min read

UAI: paper accepted (1 strong accept and 4 accepts)

This year we (Filip, Peter and I) submitted one paper to the conference on Uncertainty in Artificial Intelligence (UAI). The paper got overwhelmiingly positive feedback and will be presented virtually this summer. The current title of the paper is “99% of Distributed Optimization is a Waste of Time: The Issue and How to Fix it”, but following the reviewers’ suggestions, the title will be changed to avoid confusion about its content.

Last updated on Feb 13, 2022 1 min read

ICML reviews done

At the time when there is nothing to do, no place to travel to and no party to have fun at, reviewing papers can be a good way to kill time. I just finished my reviews for ICML, and in 2 weeks there will be another deadline, reviews for UAI. I think I have never spent so much time reviewing as with this ICML, mostly because usually there are a lot of things going on around, but not now thanks to coronavirus.

Last updated on Feb 13, 2022 1 min read

I'm one of 12 Outstanding Reviewers for AAAI

I was selected as one of 12 Outstanding Program Committee members of AAAI, out of more than 6000 reviewers in total. There were about 7737 submissions, which is 10% more than that of NeurIPS. The award will be officially given to me and the other outstanding reviewers on 11 February at 8am. I’m looking forward to going to New York for the conference!

Last updated on Feb 13, 2022 1 min read

I'm one of 12 Outstanding Reviewers for AAAI

AISTATS 2020: 3 papers accepted, 1 rejected

Today we received the final decisions for our papers submitted to the AISTATS conference (https://www.aistats.org/). Although one work was rejected, this constitutes a good acceptance rate. Unfortunately, we also had to withdraw one of our submissions. Below is the list of papers that we will present:

Revisiting Stochastic Extragradient (K. Mishchenko, D. Kovalev, E. Shulgin, Y. Malitsky and P. Richtárik)
Tighter Theory for Local SGD on Identical and Heterogeneous Data (A. Khaled, K. Mishchenko, P. Richtárik)
DAve-QN: A Distributed Averaged Quasi-Newton Method with Local Superlinear Convergence Rate (S. Soori, K. Mishchenko, A. Mokhtari, M. Dehnavi, M. Gürbüzbalaban)

Please drop by our posters in Palermo, Italy on 3-5 June!

Last updated on Feb 13, 2022 1 min read

AISTATS 2020: 3 papers accepted, 1 rejected

Paper got accepted to the SIAM Optimization journal

One of the first papers that I wrote got accepted to the SIAM journal on optimization (SIOPT). The review process was quite long and included several revisions, but I’m happy I got it accepted before my graduation. This work is a result of my collaboration with J. Malick and F. Iutzeler, from whose experience I learned a lot about optimization. The title of the work is “A Distributed Flexible Delay-tolerant Proximal Gradient Algorithm”.

Last updated on Feb 13, 2022 1 min read

I got NeurIPS Travel Award

In addition to my free NeurIPS registration, which I received as on of the top reviewers, I will also receive $1400 from the NeurIPS Foundation to sponsor my travel to the conference.

Last updated on Feb 13, 2022 1 min read

Invited talk at Boris Polyak's seminar

I was invited by Boris Polyak to present my work on Sinkhorn algorithm at his seminar. The talk took place on Tuesday, 22 October, at the Institute of Control Sciences. It was a great pleasure to hear that Boris liked my work for its simplicity. The slides of my talk are now attached to the corresponding publication (https://konstmish.github.io/publication/19_sinkhorn/).

Last updated on Feb 13, 2022 1 min read

NeurIPS workshops: 5 accepted papers

We have submitted 5 papers to 4 different workshops hosted by NeurIPS and all of them were accepted, including one work for oral presentation. The list of papers:

Stochastic Newton and Cubic Newton Methods with Simple Local Linear-Quadratic Rates (oral, with D. Kovalev and P. Richtárik)
Sinkhorn Algorithm as a Special Case of Stochastic Mirror Descent (single-author work)
Better Communication Complexity for Local SGD (with A. Khaled and P. Richtárik)
First Analysis of Local GD on Heterogeneous Data (with A. Khaled and P. Richtárik)
Revisiting Stochastic Extragradient (with D. Kovalev, E. Shulgin, Y. Malitsky and P. Richtárik)

Last updated on Feb 13, 2022 1 min read