A novel variant of the familiar backpropagation-through-time approach to training recurrent networks is described. This algorithm is intended to be used on arbitrary recurrent networks that run continually without ever being reset to an initial state, and it is specifically designed for computationally efficient computer implementation. This algorithm can be viewed as a cross between epochwise backpropagation through time, which is not appropriate for continually running networks, and the widely used on-line gradient approximation technique of truncated backpropagation through time.

Nearest neighbor classi cation assumes locally constant class conditional probabilities. This assumption becomes invalid in high dimensions with nite samples due to the curse of dimensionality. Severe bias can be introduced under these conditions when using the nearest neighbor rule. We propose a locally adaptive nearest neighbor classi cation method to try to minimize bias. We use a Chisquared distance analysis to compute a exible metric for producing neighborhoods that are highly adaptive to query locations. Neighborhoods are elongated along less relevant feature dimensions and constricted along most in uential ones. As a result, the class conditional probabilities tend to be smoother in the modi ed neighborhoods, whereby better classi cation performance can be achieved. The e cacy of our method is validated and compared against other techniques using a variety of simulated and real world data.

Any non-associative reinforcement learning algorithm can be viewed as a method for performing function optimization through (possibly noise-corrupted) sampling of function values. We describe the results of simulations in which the optima of several deterministic functions studied by Ackley were sought using variants of REINFORCE algorithms. Some of the algorithms used here incorporated additional heuristic features resembling certain aspects of some of the algorithms used in Ackley's studies. Differing levels ofperformance were achieved by the various algorithms investigated, but a number of them performed at a level comparable to the best found in Ackley's studies on a number of the tasks, in spite of their simplicity. One of these variants, called REINFORCE/MENT, represents a novel but principled approach to reinforcement learning in nontrivial networks which incorporates an entropy maximization strategy. This was found to perform especially well on more hierarchically organized tasks.

No abstract

No abstract