Phansalkar, VV and Thathachar, MAL (1995) Local and global optimization algorithms for generalized learning automata. In: Neural Computation, 7 (5). pp. 950-973.
Full text not available from this repository. (Request a copy)Abstract
Analyzes the long-term behavior of the REINFORCE and related algorithms (Williams, 1986, 1988, 1992) for generalized learning automata (Narendra and Thathachar, 1989) for the associative reinforcement learning problem (Barto and Anandan, 1985). The learning system considered is a feedforward connectionist network of generalized learning automata units. We show that REINFORCE is a gradient ascent algorithm but can exhibit unbounded behavior. A modified version of this algorithm, based on constrained optimization techniques, is suggested to overcome this disadvantage. The modified algorithm is shown to exhibit local optimization properties. A global version of the algorithm, based on constant temperature heat bath techniques, is also described and shown to converge to the global maximum. All algorithms are analyzed using weak convergence techniques
| Item Type: | Journal Article |
|---|---|
| Publication: | Neural Computation |
| Publisher: | MIT Press |
| Additional Information: | The copyright belongs to MIT Press. |
| Department/Centre: | Division of Electrical Sciences > Electrical Engineering |
| Date Deposited: | 08 Jun 2006 |
| Last Modified: | 27 Aug 2008 12:10 |
| URI: | http://eprints.iisc.ac.in/id/eprint/7557 |
Actions (login required)
 |
View Item |
