Search Results for author: Alex Olshevsky

Found 28 papers, 3 papers with code

Minimax Rate for Learning From Pairwise Comparisons in the BTL Model

no code implementations ICML 2020 Julien Hendrickx, Alex Olshevsky, Venkatesh Saligrama

We consider the problem of learning the qualities w_1, ... , w_n of a collection of items by performing noisy comparisons among them.

One-Shot Averaging for Distributed TD($λ$) Under Markov Sampling

no code implementations13 Mar 2024 Haoxing Tian, Ioannis Ch. Paschalidis, Alex Olshevsky

We consider a distributed setup for reinforcement learning, where each agent has a copy of the same Markov Decision Process but transitions are sampled from the corresponding Markov chain independently by each agent.

Convex SGD: Generalization Without Early Stopping

no code implementations8 Jan 2024 Julien Hendrickx, Alex Olshevsky

We consider the generalization error associated with stochastic gradient descent on a smooth convex function over a compact set.

On the Performance of Temporal Difference Learning With Neural Networks

no code implementations8 Dec 2023 Haoxing Tian, Ioannis Ch. Paschalidis, Alex Olshevsky

Neural Temporal Difference (TD) Learning is an approximate temporal difference method for policy evaluation that uses a neural network for function approximation.

Distributed TD(0) with Almost No Communication

no code implementations25 May 2023 Rui Liu, Alex Olshevsky

We provide a new non-asymptotic analysis of distributed temporal difference learning with linear function approximation.

Closing the gap between SVRG and TD-SVRG with Gradient Splitting

1 code implementation29 Nov 2022 Arsenii Mustafin, Alex Olshevsky, Ioannis Ch. Paschalidis

Our main result is a geometric convergence bound with predetermined learning rate of $1/8$, which is identical to the convergence bound available for SVRG in the convex setting.

A Small Gain Analysis of Single Timescale Actor Critic

no code implementations4 Mar 2022 Alex Olshevsky, Bahman Gharesifard

We consider a version of actor-critic which uses proportional step-sizes and only one critic update with a single sample from the stationary distribution per actor step.

Communication-efficient SGD: From Local SGD to One-Shot Averaging

no code implementations NeurIPS 2021 Artin Spiridonoff, Alex Olshevsky, Ioannis Ch. Paschalidis

While it is possible to obtain a linear reduction in the variance by averaging all the stochastic gradients at every step, this requires a lot of communication between the workers and the server, which can dramatically reduce the gains from parallelism.

Distributed TD(0) with Almost No Communication

no code implementations16 Apr 2021 Rui Liu, Alex Olshevsky

In the global state model, we show that the convergence rate of our distributed one-shot averaging method matches the known convergence rate of TD(0).

Temporal Difference Learning as Gradient Splitting

no code implementations27 Oct 2020 Rui Liu, Alex Olshevsky

Temporal difference learning with linear function approximation is a popular method to obtain a low-dimensional approximation of the value function of a policy in a Markov Decision Process.

Adversarial Crowdsourcing Through Robust Rank-One Matrix Completion

1 code implementation NeurIPS 2020 Qianqian Ma, Alex Olshevsky

We consider the problem of reconstructing a rank-one matrix from a revealed subset of its entries when some of the revealed entries are corrupted with perturbations that are unknown and can be arbitrarily large.

Matrix Completion

Asymptotic Convergence Rate of Alternating Minimization for Rank One Matrix Completion

no code implementations11 Aug 2020 Rui Liu, Alex Olshevsky

We study alternating minimization for matrix completion in the simplest possible setting: completing a rank-one matrix from a revealed subset of the entries.

Matrix Completion

Local SGD With a Communication Overhead Depending Only on the Number of Workers

no code implementations3 Jun 2020 Artin Spiridonoff, Alex Olshevsky, Ioannis Ch. Paschalidis

While the initial analysis of Local SGD showed it needs $\Omega ( \sqrt{T} )$ communications for $T$ local gradient steps in order for the error to scale proportionately to $1/(nT)$, this has been successively improved in a string of papers, with the state-of-the-art requiring $\Omega \left( n \left( \mbox{ polynomial in log } (T) \right) \right)$ communications.

Asymptotic Network Independence in Distributed Stochastic Optimization for Machine Learning

no code implementations28 Jun 2019 Shi Pu, Alex Olshevsky, Ioannis Ch. Paschalidis

We provide a discussion of several recent results which, in certain scenarios, are able to overcome a barrier in distributed stochastic optimization for machine learning.

BIG-bench Machine Learning Distributed Optimization

A Non-Asymptotic Analysis of Network Independence for Distributed Stochastic Gradient Descent

no code implementations6 Jun 2019 Shi Pu, Alex Olshevsky, Ioannis Ch. Paschalidis

This paper is concerned with minimizing the average of $n$ cost functions over a network, in which agents may communicate and exchange information with their peers in the network.

Graph Resistance and Learning from Pairwise Comparisons

no code implementations1 Feb 2019 Julien M. Hendrickx, Alex Olshevsky, Venkatesh Saligrama

The algorithm has a relative error decay that scales with the square root of the graph resistance, and provide a matching lower bound (up to log factors).

Robust Asynchronous Stochastic Gradient-Push: Asymptotically Optimal and Network-Independent Performance for Strongly Convex Functions

1 code implementation9 Nov 2018 Artin Spiridonoff, Alex Olshevsky, Ioannis Ch. Paschalidis

We consider the standard model of distributed optimization of a sum of functions $F(\bz) = \sum_{i=1}^n f_i(\bz)$, where node $i$ in a network holds the function $f_i(\bz)$.

Optimization and Control

Network Topology and Communication-Computation Tradeoffs in Decentralized Optimization

no code implementations26 Sep 2017 Angelia Nedić, Alex Olshevsky, Michael G. Rabbat

In decentralized optimization, nodes cooperate to minimize an overall objective function that is the sum (or average) of per-node private objective functions.

Optimization and Control Distributed, Parallel, and Cluster Computing Multiagent Systems

Crowdsourcing with Sparsely Interacting Workers

no code implementations20 Jun 2017 Yao Ma, Alex Olshevsky, Venkatesh Saligrama, Csaba Szepesvari

We then formulate a weighted rank-one optimization problem to estimate skills based on observations on an irreducible, aperiodic interaction graph.

Binary Classification Matrix Completion

Distributed Learning for Cooperative Inference

no code implementations10 Apr 2017 Angelia Nedić, Alex Olshevsky, César A. Uribe

We study the problem of cooperative inference where a group of agents interact over a network and seek to estimate a joint parameter that best explains a set of observations.

Distributed Gaussian Learning over Time-varying Directed Graphs

no code implementations6 Dec 2016 Angelia Nedić, Alex Olshevsky, César A. Uribe

We show a convergence rate of $O(1/k)$ with the constant term depending on the number of agents and the topology of the network.

A Tutorial on Distributed (Non-Bayesian) Learning: Problem, Algorithms and Results

no code implementations23 Sep 2016 Angelia Nedić, Alex Olshevsky, César A. Uribe

We overview some results on distributed learning with focus on a family of recently proposed algorithms known as non-Bayesian social learning.

Geometrically Convergent Distributed Optimization with Uncoordinated Step-Sizes

no code implementations19 Sep 2016 Angelia Nedić, Alex Olshevsky, Wei Shi, César A. Uribe

A recent algorithmic family for distributed optimization, DIGing's, have been shown to have geometric convergence over time-varying undirected/directed graphs.

Distributed Optimization

Distributed Learning with Infinitely Many Hypotheses

no code implementations6 May 2016 Angelia Nedić, Alex Olshevsky, César Uribe

We consider a distributed learning setup where a network of agents sequentially access realizations of a set of random variables with unknown distributions.

Stochastic Gradient-Push for Strongly Convex Functions on Time-Varying Directed Graphs

no code implementations9 Jun 2014 Angelia Nedic, Alex Olshevsky

We investigate the convergence rate of the recently proposed subgradient-push method for distributed optimization over time-varying directed graphs.

Optimization and Control Systems and Control

Cooperative learning in multi-agent systems from intermittent measurements

no code implementations11 Sep 2012 Naomi Ehrich Leonard, Alex Olshevsky

Motivated by the problem of tracking a direction in a decentralized way, we consider the general problem of cooperative learning in multi-agent systems with time-varying connectivity and intermittent measurements.

Distributed Subgradient Methods and Quantization Effects

no code implementations8 Mar 2008 Angelia Nedić, Alex Olshevsky, Asuman Ozdaglar, John N. Tsitsiklis

We consider a convex unconstrained optimization problem that arises in a network of agents whose goal is to cooperatively optimize the sum of the individual agent objective functions through local computations and communications.

Optimization and Control

Cannot find the paper you are looking for? You can Submit a new open access paper.