Lazy learning is a memory-based technique that, once a query is received, extractsa prediction interpolating locally the neighboring examples of the query which are considered relevant according to a distance measure. In this paper we propose a data-driven method to select on a query-by-query basis the optimal number of neighbors to be considered for each prediction. As an efficient way to identify and validate local models, the recursive least squares algorithm is introduced in the context oflocal approximation and lazy learning. Furthermore, beside the winner-takes-all strategy for model selection, a local combination of the most promising models is explored. The method proposed is tested on six different datasets and compared with a state-of-the-art approach.

To formulate a probabilistic model of this process, we consider two variables-one continuous (x), one discrete (s)-that evolve jointly in time. Thus the vector x traces out a smooth multidimensional curve, to each point of which the variable s attaches a discrete label. Markov processes on curves are based on the concept of arc length. After reviewing how to compute arc lengths along curves, we introduce a family of Markov processes whose predictions are invariant to nonlinear warpings of time. We then consider the ways in which these processes (and various generalizations) differ from HMMs. Markov Processes on Curves for Automatic Speech Recognition 753 2.1 Arc length Let g() define a D x D matrix-valued function over x E RP. If g() is everywhere nonnegative definite, then we can use it as a metric to compute distances along curves.

The Expectation-Maximization (EM) algorithm is an iterative procedure formaximum likelihood parameter estimation from data sets with missing or hidden variables [2].

H3C 3A7 CANADA 2Department of Cognitive and Neural Systems, Boston University Boston, MA 02215 USA Abstract The ARTMAP-FD neural network performs both identification (placing test patterns in classes encountered during training) and familiarity discrimination (judging whether a test pattern belongs to any of the classes encountered during training). The performance ofARTMAP-FD is tested on radar pulse data obtained in the field, and compared to that of the nearest-neighbor-based NEN algorithm and to a k 1 extension of NEN. 1 Introduction The recognition process involves both identification and familiarity discrimination. Consider, for example, a neural network designed to identify aircraft based on their radar reflections and trained on sample reflections from ten types of aircraft A . . . After training, the network should correctly classify radar reflections belonging to the familiar classes A . Familiarity discrimination is also referred to as "novelty detection," a "reject option," and "recognition in partially exposed environments."

We propose a new in-sample cross validation based method (randomized GACV) for choosing smoothing or bandwidth parameters that govern the bias-variance or fit-complexity tradeoff in'soft' classification. Soft classification refersto a learning procedure which estimates the probability that an example with a given attribute vector is in class 1 vs class O. The target for optimizing the the tradeoff is the Kullback-Liebler distance between the estimated probability distribution and the'true' probability distribution,representing knowledge of an infinite population. The method uses a randomized estimate of the trace of a Hessian and mimics cross validation at the cost of a single relearning with perturbed outcome data.

A nonlinear modification to PI control is motivated by a model of a signal transduction pathway active in mammalian blood pressure regulation.This control algorithm, labeled PII (proportional with intermittent integral), is appropriate for plants requiring exact set-pointmatching and disturbance attenuation in the presence of infrequent step changes in load disturbances or set-point. The proportional aspect of the controller is independently designed to be a disturbance attenuator and set-point matching is achieved by intermittently invoking an integral controller. The mechanisms observed in the Angiotensin 11/AT1 signaling pathway are used to control the switching of the integral control. Improved performance over PI control is shown on a model of cyc1opentenol production. A sign change in plant gain at the desirable operating point causes traditional PI control to result in an unstable system.

In robotics and other control applications it is commonplace to have a preexisting setof controllers for solving subtasks, perhaps handcrafted or previously learned or planned, and still face a difficult problem of how to choose and switch among the controllers to solve an overall task as well as possible. In this paper we present a framework based on Markov decision processes and semi-Markov decision processes for phrasing this problem, a basic theorem regarding the improvement in performance that can be obtained byswitching flexibly between given controllers, and example applications ofthe theorem. In particular, we show how an agent can plan with these high-level controllers and then use the results of such planning to find an even better plan, by modifying the existing controllers, with negligible additional cost and no re-planning. In one of our examples, the complexity of the problem is reduced from 24 billion state-action pairs to less than a million state-controller pairs. In many applications, solutions to parts of a task are known, either because they were handcrafted bypeople or because they were previously learned or planned. For example, in robotics applications, there may exist controllers for moving joints to positions, picking up objects, controlling eye movements, or navigating along hallways. More generally, an intelligent systemmay have available to it several temporally extended courses of action to choose from. In such cases, a key challenge is to take full advantage of the existing temporally extended actions,to choose or switch among them effectively, and to plan at their level rather than at the level of individual actions.

Using simulations, we show that divisive normalization is a close approximation to a maximum likelihood estimator, which, in the context of population coding, is the same as an ideal observer. We also demonstrate analytically thatthis is a general property of a large class of nonlinear recurrent networks with line attractors. Our work suggests that divisive normalization plays a critical role in noise filtering, and that every cortical layer may be an ideal observer of the activity in the preceding layer. Information processing in the cortex is often formalized as a sequence of a linear stages followed by a nonlinearity. In the visual cortex, the nonlinearity is best described bysquaring combined with a divisive pooling of local activities.

The model parameters are learned in an unsupervised manner by maximizing the likelihood that these data are generated by the model.

Since BLHT learns a stochastic model based on Bayesian Learning, the overfitting problemis reasonably solved. Moreover, BLHT has an efficient implementation. This paper shows that the model learned by BLHT converges toone which provides the most accurate predictions of percepts and rewards, given short-term memory. 1 INTRODUCTION Research on Reinforcement Learning (RL) problem forpartially observable environments is gaining more attention recently. This is mainly because the assumption that perfect and complete perception of the state of the environment is available for the learning agent, which many previous RL algorithms require, is not valid for many realistic environments.