A two step algorithm for learning from unspecific reinforcement

Details A two step algorithm for learning from unspecific reinforcement A two step algorithm for learning from unspecific reinforcement

1999-02-26

arxiv.org/abs/cond-mat/9902354v3

Physics

Condensed Matter

Statistical Mechanics

13 pages LaTeX, 4 figures, note on biologically motivated stochastic variant of the algorithm added

Scientific paper

10.1088/0305-4470/32/31/301

We study a simple learning model based on the Hebb rule to cope with "delayed", unspecific reinforcement. In spite of the unspecific nature of the information-feedback, convergence to asymptotically perfect generalization is observed, with a rate depending, however, in a non- universal way on learning parameters. Asymptotic convergence can be as fast as that of Hebbian learning, but may be slower. Moreover, for a certain range of parameter settings, it depends on initial conditions whether the system can reach the regime of asymptotically perfect generalization, or rather approaches a stationary state of poor generalization.

Affiliated with

Also associated with

No associations

Rating

A two step algorithm for learning from unspecific reinforcement does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with A two step algorithm for learning from unspecific reinforcement, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and A two step algorithm for learning from unspecific reinforcement will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-328501