Approximate and Compensate: A Method for Risk-Sensitive Meta-Deliberation and Continual Computation
MetadataShow full item record
CitationParkes, David C., and Lloyd Greenwald. 2001. Approximate and compensate: A method for risk-sensitive meta-deliberation and continual computation. In Using uncertainty within computation: Papers from the 2001 AAAI Fall Symposium: November 2-4, 2001, North Falmouth, Massachusetts, ed. C. Gomes, T. Walsh, and American Association for Artificial Intelligence, 101-108. Menlo Park, C.A.: AAAI Press.
AbstractWe present a flexible procedure for a resource-bounded agent to allocate limited computational resources to on-line problem solving. Our APPROXIMATE AND COMPENSATE methodology extends a well-known greedy time-slicing approach to conditions in which performance profiles may be non-concave and there is uncertainty in the environment and/or problem-solving procedures of an agent. With this method, the agent first approximates problem-solving performance and problem parameters with standard parameterized models. Second, the agent computes a risk-management factor that compensates for the risk inherent in the approximation. The risk-management factor represents a mean-variance tradeoff that may be derived optimally off-line using any available information. Theoretical and experimental results demonstrate that APPROXIMATE AND COMPENSATE extends existing methods to new problems and expands the practical application of meta-deliberation.
Citable link to this pagehttp://nrs.harvard.edu/urn-3:HUL.InstRepos:4101699
- FAS Scholarly Articles