Towards Interpretable Multi-Task Learning Using Bilevel Programming
This work addresses interpretability in multi-task learning for researchers and practitioners, but it appears incremental as it builds on existing sparsity and bilevel optimization techniques.
The paper tackles the problem of making multi-task learning interpretable by learning sparse graphs of task relationships, proposing a bilevel programming formulation and an efficient computational method. It shows empirically that this approach improves interpretability on synthetic and real data without sacrificing generalization performance.
Interpretable Multi-Task Learning can be expressed as learning a sparse graph of the task relationship based on the prediction performance of the learned models. Since many natural phenomenon exhibit sparse structures, enforcing sparsity on learned models reveals the underlying task relationship. Moreover, different sparsification degrees from a fully connected graph uncover various types of structures, like cliques, trees, lines, clusters or fully disconnected graphs. In this paper, we propose a bilevel formulation of multi-task learning that induces sparse graphs, thus, revealing the underlying task relationships, and an efficient method for its computation. We show empirically how the induced sparse graph improves the interpretability of the learned models and their relationship on synthetic and real data, without sacrificing generalization performance. Code at https://bit.ly/GraphGuidedMTL