We describe a computer system that provides a real-time musical accompanimentfor a live soloist in a piece of non-improvised music for soloist and accompaniment. A Bayesian network is developed thatrepresents the joint distribution on the times at which the solo and accompaniment notes are played, relating the two parts through a layer of hidden variables. The network is first constructed usingthe rhythmic information contained in the musical score. The network is then trained to capture the musical interpretations ofthe soloist and accompanist in an off-line rehearsal phase. During live accompaniment the learned distribution of the network is combined with a real-time analysis of the soloist's acoustic signal, performedwith a hidden Markov model, to generate a musically principledaccompaniment that respects all available sources of knowledge. A live demonstration will be provided.

Discriminative classifiers model the posterior p(ylx)directly, or learn a direct map from inputs x to the class labels. There are several compelling reasons for using discriminative rather than generative classifiers, oneof which, succinctly articulated by Vapnik [6], is that "one should solve the [classification] problem directly and never solve a more general problem as an intermediate step [such as modeling p(xly)]." Indeed, leaving aside computational issues and matters such as handling missing data, the prevailing consensus seems to be that discriminative classifiers are almost always to be preferred to generative ones. Anotherpiece of prevailing folk wisdom is that the number of examples needed to fit a model is often roughly linear in the number of free parameters of a model. This has its theoretical basis in the observation that for "many" models, the VC dimension is roughly linear or at most some low-order polynomial in the number of parameters (see, e.g., [1, 3]), and it is known that sample complexity in the discriminative setting is linear in the VC dimension [6]. In this paper, we study empirically and theoretically the extent to which these beliefs are true. A parametric family of probabilistic models p(x, y) can be fit either to optimize the joint likelihood of the inputs and the labels, or fit to optimize the conditional likelihood p(ylx), or even fit to minimize the 0-1 training error obtained by thresholding p(ylx) to make predictions.

We are interested in the mechanisms by which individuals monitor and adjust their performance of simple cognitive tasks. We model a speeded discrimination task in which individuals are asked to classify a sequence of stimuli (Jones & Braver, 2001). Response conflict arises when one stimulus class is infrequent relative to another, resulting in more errors and slower reaction times for the infrequent class. How do control processes modulatebehavior based on the relative class frequencies? We explain performance from a rational perspective that casts the goal of individuals as minimizing a cost that depends both on error rate and reaction time.With two additional assumptions of rationality--that class prior probabilities are accurately estimated and that inference is optimal subject to limitations on rate of information transmission--we obtain a good fit to overall RT and error data, as well as trial-by-trial variations in performance.

In previous work on "transformed mixtures of Gaussians" and "transformed hidden Markov models", we showed how the EM algorithm ina discrete latent variable model can be used to jointly normalize data (e.g., center images, pitch-normalize spectrograms) and learn a mixture model of the normalized data. The only input to the algorithm is the data, a list of possible transformations, and the number of clusters to find. The main criticism of this work was that the exhaustive computation of the posterior probabilities overtransformations would make scaling up to large feature vectors and large sets of transformations intractable. Here, we describe howa tremendous speedup is acheived through the use of a variational technique for decoupling transformations, and a fast Fourier transform method for computing posterior probabilities.

We consider the use of two additive control variate methods to reduce the variance of performance gradient estimates in reinforcement learning problems.The first approach we consider is the baseline method, in which a function of the current state is added to the discounted value estimate. We relate the performance of these methods, which use sample paths,to the variance of estimates based on iid data. We derive the baseline function that minimizes this variance, and we show that the variance forany baseline is the sum of the optimal variance and a weighted squared distance to the optimal baseline. We show that the widely used average discounted value baseline (where the reward is replaced by the difference between the reward and its expectation) is suboptimal. The second approach we consider is the actor-critic method, which uses an approximate value function. We give bounds on the expected squared error of its estimates. We show that minimizing distance to the true value function is suboptimal in general; we provide an example for which the true value function gives an estimate with positive variance, but the optimal valuefunction gives an unbiased estimate with zero variance. Our bounds suggest algorithms to estimate the gradient of the performance of parameterized baseline or value functions.

We present an algorithm that induces a class of models with thin junction trees--models that are characterized by an upper bound on the size of the maximal cliques of their triangulated graph. By ensuring that the junction tree is thin, inference in our models remains tractable throughout the learning process. This allows both an efficient implementation of an iterative scaling parameter estimation algorithm and also ensures that inference can be performed efficiently with the final model. We illustrate the approach with applications in handwritten digit recognition and DNA splice site detection.

Statistical learning and probabilistic inference techniques are used to infer thehand position of a subject from multi-electrode recordings of neural activityin motor cortex. First, an array of electrodes provides training dataof neural firing conditioned on hand kinematics. We learn a nonparametric representationof this firing activity using a Bayesian model and rigorously compare it with previous models using cross-validation. Second, we infer a posterior probability distribution over hand motion conditioned on a sequence of neural test data using Bayesian inference. The learned firing models of multiple cells are used to define a non-Gaussian likelihood term which is combined with a prior probability for the kinematics. A particle filtering method is used to represent, update, and propagate the posterior distribution over time. The approach is compared withtraditional linear filtering methods; the results suggest that it may be appropriate for neural prosthetic applications.

It is well known that under noisy conditions we can hear speech much more clearly when we read the speaker's lips. This suggests theutility of audiovisual information for the task of speech enhancement. We propose a method to exploit audiovisual cues to enable speech separation under non-stationary noise and with a single microphone. We revise and extend HMM-based speech enhancement techniques, in which signal and noise models are factori allycombined, to incorporate visual lip information and employ novelsignal HMMs in which the dynamics of narrow-band and wide band components are factorial. We avoid the combinatorial explosionin the factorial model by using a simple approximate inference technique to quickly estimate the clean signals in a mixture. We present a preliminary evaluation of this approach using a small-vocabulary audiovisual database, showing promising improvements in machine intelligibility for speech enhanced using audio and visual information.

We describe the application of kernel methods to Natural Language Processing (NLP)problems. In many NLP tasks the objects being modeled are strings, trees, graphs or other discrete structures which require some mechanism to convert them into feature vectors. We describe kernels for various natural language structures, allowing rich, high dimensional representations ofthese structures. We show how a kernel over trees can be applied to parsing using the voted perceptron algorithm, and we give experimental results on the ATIS corpus of parse trees.

The Temporal Coding Hypothesis of Miller and colleagues [7] suggests thatanimals integrate related temporal patterns of stimuli into single memory representations. We formalize this concept using quasi-Bayes estimation to update the parameters of a constrained hiddenMarkov model. This approach allows us to account for some surprising temporal effects in the second order conditioning experimentsof Miller et al. [1, 2, 3], which other models are unable to explain. 1 Introduction Animal learning involves more than just predicting reinforcement. The well-known phenomena of latent learning and sensory preconditioning indicate that animals learn about stimuli in their environment before any reinforcement is supplied. More recently, a series of experiments by R. R. Miller and colleagues has demonstrated that in classical conditioning paradigms, animals appear to learn the temporal structure ofthe stimuli [8].