Bayesian Multitask Learning with Latent Hierarchies

TL;DR

This paper introduces a hierarchical Bayesian model for multitask learning that captures latent relationships between tasks, effectively unifying various existing models and demonstrating strong performance on real-world datasets.

Contribution

It proposes a novel hierarchical Bayesian framework that models latent task relationships, unifying and extending prior multitask learning approaches.

Findings

Model subsumes several existing multitask learning models.

Performs well on three real-world datasets.

Effectively captures latent task hierarchies.

Abstract

We learn multiple hypotheses for related tasks under a latent hierarchical relationship between tasks. We exploit the intuition that for domain adaptation, we wish to share classifier structure, but for multitask learning, we wish to share covariance structure. Our hierarchical model is seen to subsume several previously proposed multitask learning models and performs well on three distinct real-world data sets.