Discovering Structure in Multiple Learning Tasks: The TC Algorithm

Recently, there has been an increased interest in "lifelong" machine learning methods, that transfer knowledge across multiple learning tasks. Such methods have repeatedly been found to outperform conventional, single-task learningalgorithms when the learningtasks are appropriatelyrelated. To increase robustness of such approaches, methods are desirable that can reason about the relatedness of individuallearning tasks, in order to avoid the danger arising from tasks that are unrelated and thus potentially misleading. This paper describes the task-clustering (TC) algorithm. TC clusters learning tasks into classes of mutually related tasks. When facing a new learning task, TC first determines the most related task cluster, then exploits information selectively from this task cluster only. An empirical study carried out in a mobile robot domain shows that TC outperforms its non-selective counterpart in situations where only a small number of tasks is relevant.

• Type: Conference
• Year: 1996
• Where: ICML
• Authors: Sebastian Thrun, Joseph O'Sullivan
• Keywords: ICML 1996, Learning Task, Machine Learning, Multiple Learning Tasks, Related Task