Algorithms for Learning and Decision Making
Abstract
We have investigated learning algorithms for inference and decision making, by using exact and approximate optimization methods. Most of our research has been in approximate dynamic programming/reinforcement learning methods, with a focus on Markovian Decision Problems with a very large number of states. Much of our work is related to a fundamental algorithm, Q-learning, and related new methods that relate to exact and approximate policy iteration. In particular, we have investigated, convergence issues, error bounds, policy oscillation, exploration-enhanced methods, and issues of decision making in a multi-agent environment. Another research area is large-scale convex optimization methods, with a focus on problems whose cost function involves a sum of a large number of component functions. This includes a unifying framework for polyhedral approximation recently proposed by the principal investigator, incremental gradient and subgradient methods, which are currently at the forefront of algorithmic machine learning research, as well as a new incremental version of the proximal minimization algorithm. We have developed new polyhedral approximation algorithms, including a simplicial decomposition method that applies to large-scale conic programming problems.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 09, 2013
- Accession Number
- ADA591909
Entities
People
- Dimitri P. Bertsekas
Organizations
- Massachusetts Institute of Technology