no code implementations • 1 Dec 2023 • Noga Alon, Dmitrii Avdiukhin, Dor Elboim, Orr Fischer, Grigory Yaroslavtsev
Contrastive learning is a highly successful technique for learning representations of data from labeled tuples, specifying the distance relations within the tuple.
no code implementations • 9 Feb 2023 • Dmitrii Avdiukhin, Grigory Yaroslavtsev, Danny Vainstein, Orr Fischer, Sauman Das, Faraz Mirza
Consider a collection of data tuples labeled according to their hierarchical structure.
no code implementations • 27 May 2022 • Dmitrii Avdiukhin, Grigory Yaroslavtsev
We give the first polynomial time algorithms for escaping from high-dimensional saddle points under a moderate number of constraints.
no code implementations • NeurIPS 2021 • Dmitrii Avdiukhin, Grigory Yaroslavtsev
Gradient compression methods can be used to alleviate this problem, and a recent line of work shows that SGD augmented with gradient compression converges to an $\varepsilon$-first-order stationary point.
no code implementations • 15 Dec 2020 • Stanislav Naumov, Grigory Yaroslavtsev, Dmitrii Avdiukhin
In order to address the challenge of scaling up hierarchical clustering to such large datasets we propose a new practical hierarchical clustering algorithm B++&C.
1 code implementation • 12 Oct 2019 • Dmitrii Avdiukhin, Grigory Yaroslavtsev, Samson Zhou
Our analysis is based on a new set of first-order linear differential inequalities and their robust approximate versions.
no code implementations • 7 May 2019 • Dmitrii Avdiukhin, Slobodan Mitrović, Grigory Yaroslavtsev, Samson Zhou
We propose the first adversarially robust algorithm for monotone submodular maximization under single and multiple knapsack constraints with scalable implementations in distributed and streaming settings.
no code implementations • ICML 2018 • Grigory Yaroslavtsev, Adithya Vadapalli
We present first massively parallel (MPC) algorithms and hardness of approximation results for computing Single-Linkage Clustering of n input d-dimensional vectors under Hamming, $\ell_1, \ell_2$ and $\ell_\infty$ distances.
no code implementations • 10 Aug 2012 • Sofya Raskhodnikova, Grigory Yaroslavtsev
We show that an analog of Hastad's switching lemma holds for pseudo-Boolean k-DNFs if all constants associated with the terms of the formula are bounded.