Compressive Feature Selection for Remote Visual Multi-Task Inference
This work addresses feature compression for remote multi-task inference, which is incremental as it builds on existing methods by applying mutual information in a novel way.
The paper tackled the problem of determining feature importance for multi-task inference in deep models, particularly for remote visual applications, and found that mutual information between features and task outputs is an effective measure, with experiments showing it outperforms alternative approaches in both hard and soft selection scenarios.
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.