Compressive Feature Selection for Remote Visual Multi-Task Inference

TL;DR

This paper investigates the effectiveness of mutual information as a measure of feature importance in multi-task remote inference, proposing compression methods based on MI and comparing them with alternatives.

Contribution

It introduces MI-based feature importance assessment for multi-task models and evaluates its effectiveness against other selection approaches.

Findings

MI can effectively measure feature importance for multi-task inference

MI-based compression outperforms some alternative feature selection methods

Multi-objective analysis provides deeper insights into feature importance trade-offs

Abstract

Deep models produce a number of features in each internal layer. A key problem in applications such as feature compression for remote inference is determining how important each feature is for the task(s) performed by the model. The problem is especially challenging in the case of multi-task inference, where the same feature may carry different importance for different tasks. In this paper, we examine how effective is mutual information (MI) between a feature and a model's task output as a measure of the feature's importance for that task. Experiments involving hard selection and soft selection (unequal compression) based on MI are carried out to compare the MI-based method with alternative approaches. Multi-objective analysis is provided to offer further insight.