Do Outliers Ruin Collaboration?

05/12/2018
by   Mingda Qiao, et al.
0

We consider the problem of learning a binary classifier from n different data sources, among which at most an η fraction are adversarial. The overhead is defined as the ratio between the sample complexity of learning in this setting and that of learning the same hypothesis class on a single data distribution. We present an algorithm that achieves an O(η n + n) overhead, which is proved to be worst-case optimal. We also discuss the potential challenges to the design of a computationally efficient learning algorithm with a small overhead.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
11/11/2021

A 2-Dimensional Binary Search for Integer Pareto Frontiers

For finite integer squares, we consider the problem of learning a classi...
research
05/23/2018

Tight Bounds for Collaborative PAC Learning via Multiplicative Weights

We study the collaborative PAC learning problem recently proposed in Blu...
research
12/13/2012

Learning Sparse Low-Threshold Linear Classifiers

We consider the problem of learning a non-negative linear classifier wit...
research
03/05/2023

Improved Sample Complexity Bounds for Distributionally Robust Reinforcement Learning

We consider the problem of learning a control policy that is robust agai...
research
02/24/2020

On the Sample Complexity of Adversarial Multi-Source PAC Learning

We study the problem of learning from multiple untrusted data sources, a...
research
03/25/2021

Active Structure Learning of Bayesian Networks in an Observational Setting

We study active structure learning of Bayesian networks in an observatio...
research
10/28/2020

Online feature selection for rapid, low-overhead learning in networked systems

Data-driven functions for operation and management often require measure...

Please sign up or login with your details

Forgot password? Click here to reset