Current browse context:
stat.ML
Change to browse by:
References & Citations
Statistics > Machine Learning
Title: Balancing New Against Old Information: The Role of Surprise in Learning
(Submitted on 17 Jun 2016 (v1), last revised 1 Mar 2017 (this version, v2))
Abstract: Surprise describes a range of phenomena from unexpected events to behavioral responses. We propose a measure of surprise and use it for surprise-driven learning. Our surprise measure takes into account data likelihood as well as the degree of commitment to a belief via the entropy of the belief distribution. We find that surprise-minimizing learning dynamically adjusts the balance between new and old information without the need of knowledge about the temporal statistics of the environment. We apply our framework to a dynamic decision-making task and a maze exploration task. Our surprise minimizing framework is suitable for learning in complex environments, even if the environment undergoes gradual or sudden changes and could eventually provide a framework to study the behavior of humans and animals encountering surprising events.
Submission history
From: Mohammad Javad Faraji [view email][v1] Fri, 17 Jun 2016 19:54:43 GMT (2340kb,D)
[v2] Wed, 1 Mar 2017 20:31:24 GMT (2057kb,D)
Link back to: arXiv, form interface, contact.