Compressed Hierarchical Representations for Multi-Task Learning and Task Clustering

TL;DR

This paper introduces a hierarchical representation learning framework for multi-task learning that uses information bottleneck principles to produce disentangled, interpretable task-agnostic and task-specific features, achieving competitive results.

Contribution

It proposes a novel hierarchical representation approach for MTL based on information bottleneck and additive noise models, enabling task similarity analysis and interpretability.

Findings

Achieves competitive performance on MTL benchmarks.

Task similarity can be inferred from trained noise model parameters.

Produces disentangled, interpretable representations.

Abstract

In this paper, we frame homogeneous-feature multi-task learning (MTL) as a hierarchical representation learning problem, with one task-agnostic and multiple task-specific latent representations. Drawing inspiration from the information bottleneck principle and assuming an additive independent noise model between the task-agnostic and task-specific latent representations, we limit the information contained in each task-specific representation. It is shown that our resulting representations yield competitive performance for several MTL benchmarks. Furthermore, for certain setups, we show that the trained parameters of the additive noise model are closely related to the similarity of different tasks. This indicates that our approach yields a task-agnostic representation that is disentangled in the sense that its individual dimensions may be interpretable from a task-specific perspective.