Online Learning in Monkeys

We examine online learning in the context of the Wisconsin Card Sorting Task (WCST), a task for which the concept acquisition strategies for human and other primates are well documented. We describe a new WCST experiment in rhesus monkeys, comparing the monkeys' behaviors to that of online learning algorithms. Our expectation is that insights gained from this work and future research can lead to improved artificial learning systems. WCST as Online Learning Online learning (Blum 1998; Kivinen, Smola, and Williamson 2004) assumes that an adversary provides data sequentially to a learner. In particular, the adversary can change the target concept at any time (called a concept drift), unbeknownst to the learner. This is in sharp contrast to the stationary independent and identically-distributed (iid) data assumption in standard learning models. As such, online learning naturally models changing environments. An interesting question is how intelligent animals learn under similar condi...

• Type: Conference
• Year: 2008
• Where: AAAI
• Authors: Xiaojin Zhu, Michael Coen, Shelley Prudom, Ricki Colman, Joseph Kemnitz
• Keywords: AAAI 2008, Intelligent Agents, Online Learning, Online Learning Algorithms, WCST