
Personalized Federated Learning for Heterogeneous Clients with Clustered Knowledge Transfer
Personalized federated learning (FL) aims to train model(s) that can per...
read it

BestArm Identification in Correlated MultiArmed Bandits
In this paper we consider the problem of bestarm identification in mult...
read it

Job Dispatching Policies for Queueing Systems with Unknown Service Rates
In multiserver queueing systems where there is no central queue holding...
read it

Local Adaptivity in Federated Learning: Convergence and Consistency
The federated learning (FL) framework trains a machine learning model us...
read it

Adaptive Quantization of Model Updates for CommunicationEfficient Federated Learning
Communication of model updates between client nodes and the central aggr...
read it

Synergy via Redundancy: Adaptive Replication Strategies and Fundamental Limits
The maximum possible throughput (or the rate of job completion) of a mul...
read it

Banditbased CommunicationEfficient Client Selection Strategies for Federated Learning
Due to communication constraints and intermittent client availability in...
read it

Client Selection in Federated Learning: Convergence Analysis and PowerofChoice Selection Strategies
Federated learning is a distributed optimization paradigm that enables a...
read it

Service Rate Region: A New Aspect of Coded Distributed System Design
Erasure coding has been recently employed as a powerful method to mitiga...
read it

Tackling the Objective Inconsistency Problem in Heterogeneous Federated Optimization
In federated optimization, heterogeneity in the clients' local datasets ...
read it

Slow and Stale Gradients Can Win the Race
Distributed Stochastic Gradient Descent (SGD) when run in a synchronous ...
read it

Machine Learning on Volatile Instances
Due to the massive size of the neural network models and training datase...
read it

Overlap LocalSGD: An Algorithmic Approach to Hide Communication Delays in Distributed SGD
Distributed stochastic gradient descent (SGD) is essential for scaling t...
read it

Advances and Open Problems in Federated Learning
Federated learning (FL) is a machine learning setting where many clients...
read it

MultiArmed Bandits with Correlated Arms
We consider a multiarmed bandit framework where the rewards obtained by...
read it

Deep Probabilistic Kernels for SampleEfficient Learning
Gaussian Processes (GPs) with an appropriate kernel are known to provide...
read it

Accelerating Deep Learning by Focusing on the Biggest Losers
This paper introduces SelectiveBackprop, a technique that accelerates t...
read it

MATCHA: Speeding Up Decentralized SGD via Matching Decomposition Sampling
The tradeoff between convergence error and communication delays in dece...
read it

SysML: The New Frontier of Machine Learning Systems
Machine learning (ML) techniques are enjoying rapidly increasing adoptio...
read it

Service Rate Region of Content Access from Erasure Coded Storage
We consider storage systems in which K files are stored over N nodes. A ...
read it

Adaptive Communication Strategies to Achieve the Best ErrorRuntime Tradeoff in LocalUpdate SGD
Largescale machine learning training, in particular distributed stochas...
read it

Exploiting Correlation in FiniteArmed Structured Bandits
We consider a correlated multiarmed bandit problem in which rewards of ...
read it

Cooperative SGD: A unified Framework for the Design and Analysis of CommunicationEfficient SGD Algorithms
Stateoftheart distributed machine learning suffers from significant d...
read it

Correlated Multiarmed Bandits with a Latent Random Source
We consider a novel multiarmed bandit framework where the rewards obtai...
read it

Active Distribution Learning from Indirect Samples
This paper studies the problem of learning the probability distribution...
read it

Rateless Codes for NearPerfect Load Balancing in Distributed MatrixVector Multiplication
Largescale machine learning and data mining applications require comput...
read it

Slow and Stale Gradients Can Win the Race: ErrorRuntime Tradeoffs in Distributed SGD
Distributed Stochastic Gradient Descent (SGD) when run in a synchronous ...
read it
Gauri Joshi
is this you? claim profile