Using Task Descriptions in Lifelong Machine Learning for Improved Performance and Zero-Shot Transfer

TL;DR

This paper introduces a lifelong learning approach that leverages high-level task descriptions to model inter-task relationships, enhancing performance and enabling zero-shot transfer without additional training data.

Contribution

It proposes a coupled dictionary learning method that uses task descriptors for better knowledge transfer and zero-shot learning in lifelong machine learning.

Findings

Improves task policy performance using task descriptions.

Provides theoretical justification for descriptor-based transfer.

Enables zero-shot learning for new tasks.

Abstract

Knowledge transfer between tasks can improve the performance of learned models, but requires an accurate estimate of the inter-task relationships to identify the relevant knowledge to transfer. These inter-task relationships are typically estimated based on training data for each task, which is inefficient in lifelong learning settings where the goal is to learn each consecutive task rapidly from as little data as possible. To reduce this burden, we develop a lifelong learning method based on coupled dictionary learning that utilizes high-level task descriptions to model the inter-task relationships. We show that using task descriptors improves the performance of the learned task policies, providing both theoretical justification for the benefit and empirical demonstration of the improvement across a variety of learning problems. Given only the descriptor for a new task, the lifelong learner is also able to accurately predict a model for the new task through zero-shot learning using the coupled dictionary, eliminating the need to gather training data before addressing the task.