Simulationresults show that it is efficient and precise, when compared with cross-validation and other techniques which require matrix inversion.

I derive approximations to the learning curves for the more generic case of mismatched models, and find very rich behaviour: For large input space dimensionality, where the results become exact, there are universal (student-independent) plateaux in the learning curve, with transitions in between that can exhibit arbitrarily many over-fitting maxima; over-fitting can occur even if the student estimates the teacher noise level correctly. In lower dimensions, plateaux also appear, and the learning curve remains dependent on the mismatch between student and teacher even in the asymptotic limit of a large number of training examples. Learning withexcessively strong smoothness assumptions can be particularly dangerous:For example, a student with a standard radial basis function covariance function will learn a rougher teacher function onlylogarithmically slowly. All predictions are confirmed by simulations. 1 Introduction There has in the last few years been a good deal of excitement about the use of Gaussian processes (GPs) as an alternative to feedforward networks [1]. GPs make prior assumptions about the problem to be learned very transparent, and even though they are nonparametric models, inference-at least in the case of regression considered below-is relatively straightforward. One crucial question for applications is then how'fast' GPs learn, i.e. how many training examples are needed to achieve a certain level of generalization performance.

To find out how the representations of structured visual objects depend on the co-occurrence statistics of their constituents, we exposed subjects to a set of composite images with tight control exerted over (1) the conditional probabilitiesof the constituent fragments, and (2) the value of Barlow's criterion of "suspicious coincidence" (the ratio of joint probability to the product of marginals). We then compared the part verification response timesfor various probe/target combinations before and after the exposure. For composite probes, the speedup was much larger for targets thatcontained pairs of fragments perfectly predictive of each other, compared to those that did not. This effect was modulated by the significance oftheir co-occurrence as estimated by Barlow's criterion. For lone-fragment probes, the speedup in all conditions was generally lower than for composites. These results shed light on the brain's strategies for unsupervised acquisition of structural information in vision.

Singularities are ubiquitous in the parameter space of hierarchical models such as multilayer perceptrons. At singularities, the Fisher information matrix degenerates, and the Cramer-Rao paradigm does no more hold, implying that the classical model selection theory suchas AIC and MDL cannot be applied. It is important to study the relation between the generalization error and the training error at singularities. The present paper demonstrates a method of analyzing these errors both for the maximum likelihood estimator andthe Bayesian predictive distribution in terms of Gaussian random fields, by using simple models. 1 Introduction A neural network is specified by a number of parameters which are synaptic weights and biases. Learning takes place by modifying these parameters from observed input-output examples.

A new class of Support Vector Machine (SVM) that is applicable to sequential-pattern recognition such as speech recognition is developed by incorporating an idea of nonlinear time alignment into the kernel function. Since the time-alignment operation of sequential pattern is embedded in the new kernel function, standard SVM training and classification algorithms can be employed without further modifications. The proposed SVM (DTAK-SVM) is evaluated in speaker-dependent speech recognition experiments of hand-segmented phoneme recognition. Preliminary experimental results show comparable recognition performance with hidden Markov models (HMMs).

The information bottleneck method is an unsupervised model independent data organization technique. Given a joint distribution peA, B), this method constructs anew variable T that extracts partitions, or clusters, over the values of A that are informative about B. In a recent paper, we introduced a general principled frameworkfor multivariate extensions of the information bottleneck method that allows us to consider multiple systems of data partitions that are interrelated. In this paper, we present a new family of simple agglomerative algorithms to construct such systems of interrelated clusters. We analyze the behavior of these algorithms and apply them to several real-life datasets.

We consider the problem of learning to attain multiple goals in a dynamic environment, whichis initially unknown. In addition, the environment may contain arbitrarily varying elements related to actions of other agents or to non-stationary moves of Nature. This problem is modelled as a stochastic (Markov) game between the learning agent and an arbitrary player, with a vector-valued reward function. The objective of the learning agent is to have its long-term average reward vector belong to a given target set. We devise an algorithm for achieving this task, which is based on the theory of approachability for stochastic games. This algorithm combines, inan appropriate way, a finite set of standard, scalar-reward learning algorithms. Sufficientconditions are given for the convergence of the learning algorithm to a general target set. The specialization of these results to the single-controller Markov decision problem are discussed as well.

The first is the widely used optimistic Q-learning, which initializes the Q-values to large initial values and then follows a greedy policy with respect to the Q-values. We show that setting the initial value sufficiently large guarantees the converges to an E optimal policy. The second is a new and novel algorithm incremental Q-learning,which gradually promotes the values of actions that are not taken. We show that incremental Q-learning converges, in the limit, to the optimal policy. Our incremental Q-learning algorithm canbe viewed as derandomization of the E-greedy Q-learning. 1 Introduction One of the challenges of Reinforcement Learning is learning in an unknown environment.

There have been active research efforts in this area.

In the'missing data' approach to improving the robustness of automatic speech recognition to added noise, an initial process identifies spectraltemporal regionswhich are dominated by the speech source. The remaining regions are considered to be'missing'. In this paper we develop a connectionist approach to the problem of adapting speech recognition to the missing data case, using Recurrent Neural Networks. In contrast to methods based on Hidden Markov Models, RNNs allow us to make use of long-term time constraints and to make the problems of classification with incomplete data and imputing missing values interact. We report encouraging results on an isolated digit recognition task.