One network fits all? Modular versus monolithic task formulations in neural networks

Abhimanyu Das
Atish Agarwala
Brendan Juba
Vatsal Sharan
ICLR 2021 (2021)
Google Scholar

Abstract

Can deep learning solve multiple, very different tasks simultaneously? We investigate how the representations of the underlying tasks affect the ability of a single neural network to learn them jointly. We present theoretical and empirical findings that a single neural network is capable of simultaneously learning multiple tasks from a combined data set, for a variety of methods for representing tasks---for example, when the distinct tasks are represented by well-separated clusters or decision trees over some task-code attributes. Indeed, more strongly, we present a novel analysis that shows that families of simple programming-like constructs for the task codings are learnable by two-layer neural networks with standard training. We study more generally how the complexity of learning such combined tasks grows with the complexity of the task codes; we find that learning many tasks can be provably hard, even though the individual tasks are easy to learn. We provide empirical support for the usefulness of the learning bounds by training networks on clusters, decision trees, and SQL-style aggregation.