Abstract | ||
---|---|---|
Surprise is a widely used concept describing a range of phenomena from unexpected events to behavioral responses. We propose a measure of surprise, to arrive at a new framework for surprise-driven learning. There are two components to this framework: (i) a confidence-adjusted surprise measure to capture environmental statistics as well as subjective beliefs, (ii) a surprise-minimization learning rule, or SMiLe-rule, which dynamically adjusts the balance between new and old information without making prior assumptions about the temporal statistics of the environment. We apply our framework to a dynamic decision-making task and a maze exploration task to demonstrate that it is suitable for learning in complex environments, even if the environment undergoes gradual or sudden changes. Our proposed surprise-modulated belief update algorithm provides a framework to study the behavior of humans and animals encountering surprising events. |
Year | Venue | Field |
---|---|---|
2016 | arXiv: Machine Learning | Environmental statistics,Learning rule,Artificial intelligence,Surprise,Unexpected events,Mathematics,Machine learning |
DocType | Volume | Citations |
Journal | abs/1606.05642 | 0 |
PageRank | References | Authors |
0.34 | 4 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Mohammad Javad Faraji | 1 | 57 | 3.69 |
Kerstin Preuschoff | 2 | 20 | 3.23 |
Wulfram Gerstner | 3 | 2437 | 410.08 |