Network Classifiers With Output Smoothing

TL;DR

This paper proposes output smoothing strategies for network classifiers with heterogeneous agents, enabling cooperation despite varying feature and classifier dimensions, and demonstrates their effectiveness through simulations.

Contribution

Introduces two output smoothing strategies for heterogeneous network classifiers, addressing the challenge of differing classifier dimensions among agents.

Findings

Global and local smoothing improve classification accuracy

Output smoothing facilitates cooperation among heterogeneous agents

Simulation results validate the proposed methods' effectiveness

Abstract

This work introduces two strategies for training network classifiers with heterogeneous agents. One strategy promotes global smoothing over the graph and a second strategy promotes local smoothing over neighbourhoods. It is assumed that the feature sizes can vary from one agent to another, with some agents observing insufficient attributes to be able to make reliable decisions on their own. As a result, cooperation with neighbours is necessary. However, due to the fact that the feature dimensions are different across the agents, their classifier dimensions will also be different. This means that cooperation cannot rely on combining the classifier parameters. We instead propose smoothing the outputs of the classifiers, which are the predicted labels. By doing so, the dynamics that describes the evolution of the network classifier becomes more challenging than usual because the classifier parameters end up appearing as part of the regularization term as well. We illustrate performance by means of computer simulations.