We derive multiplicative updates for solving the nonnegative quadratic programming problem in support vector machines (SVMs). The updates have a simple closed form, and we prove that they converge monotonically tothe solution of the maximum margin hyperplane. The updates optimize the traditionally proposed objective function for SVMs. They do not involve any heuristics such as choosing a learning rate or deciding which variables to update at each iteration. They can be used to adjust all the quadratic programming variables in parallel with a guarantee of improvement ateach iteration. We analyze the asymptotic convergence of the updates and show that the coefficients of nonsupport vectors decay geometrically to zero at a rate that depends on their margins.

We present a novel, flexible statistical approach for modelling music and text jointly. The approach is based on multi-modal mixture models and maximum a posteriori estimation using EM. The learned models can be used to browse databases with documents containing music and text, to search for music using queries consisting of music and text (lyrics and other contextual information), to annotate text documents with music, and to automatically recommend or identify similar songs.

We present a simple direct approach for solving the ICA problem, using density estimation and maximum likelihood. Given a candidate orthogonalframe, we model each of the coordinates using a semi-parametric density estimate based on cubic splines. Since our estimates have two continuous derivatives, we can easily run a second ordersearch for the frame parameters. Our method performs very favorably when compared to state-of-the-art techniques. 1 Introduction Independent component analysis (ICA) is a popular enhancement over principal component analysis (PCA) and factor analysis. IRP which is assumed to arise from a linear mixing of a latent random source vector S E IRP, (1) X AS; the components Sj, j 1, ...,p of S are assumed to be independently distributed.

We introduce a class of string kernels, called mismatch kernels, for use with support vector machines (SVMs) in a discriminative approach to the protein classification problem.

We develop and test new machine learning methods for the prediction oftopological representations of protein structures in the form of coarse-or fine-grained contact or distance maps that are translation androtation invariant. The methods are based on generalized input-output hidden Markov models (GIOHMMs) and generalized recursive neural networks (GRNNs). The methods are used to predict topologydirectly in the fine-grained case and, in the coarsegrained case,indirectly by first learning how to score candidate graphs and then using the scoring function to search the space of possible configurations. Computer simulations show that the predictors achievestate-of-the-art performance.

We establish a new hardness result that shows that the difficulty of planning infactored Markov decision processes is representational rather than just computational. More precisely, we give a fixed family of factored MDPswith linear rewards whose optimal policies and value functions simply cannot be represented succinctly in any standard parametric form. Previous hardness results indicated that computing good policies from the MDP parameters was difficult, but left open the possibility of succinct function approximation for any fixed factored MDP. Our result applies even to policies which yield a polynomially poor approximation to the optimal value, and highlights interesting connections with the complexity classof Arthur-Merlin games.

We propose a dynamic Bayesian model for motifs in biopolymer sequences whichcaptures rich biological prior knowledge and positional dependencies in motif structure in a principled way. Our model posits that the position-specific multinomial parameters for monomer distribution aredistributed as a latent Dirichlet-mixture random variable, and the position-specific Dirichlet component is determined by a hidden Markov process. Model parameters can be fit on training motifs using a variational EMalgorithm within an empirical Bayesian framework. Variational inference is also used for detecting hidden motifs. Our model improves overprevious models that ignore biological priors and positional dependence. It has much higher sensitivity to motifs during detection and a notable ability to distinguish genuine motifs from false recurring patterns.

We present a hierarchical Bayesian model for learning efficient codes of higher-order structure in natural images. The model, a nonlinear generalization ofindependent component analysis, replaces the standard assumption of independence for the joint distribution of coefficients with a distribution that is adapted to the variance structure of the coefficients of an efficient image basis. This offers a novel description of higherorder imagestructure and provides a way to learn coarse-coded, sparsedistributed representationsof abstract image properties such as object location, scale, and texture.

Modeling the distributions of the independent sources with mixture of Gaussians allows sources to be estimated with different kurtosis and skewness. The variational Bayesian method automatically determines the dimensionality of the data and yields an accurate density model for the observed data without overfitting problems.

We propose a model for natural images in which the probability of an image isproportional to the product of the probabilities of some filter outputs. Weencourage the system to find sparse features by using a Studentt distribution to model each filter output. If the t-distribution is used to model the combined outputs of sets of neurally adjacent filters, the system learnsa topographic map in which the orientation, spatial frequency and location of the filters change smoothly across the map. Even though maximum likelihood learning is intractable in our model, the product form allows a relatively efficient learning procedure that works well even for highly overcomplete sets of filters. Once the model has been learned it can be used as a prior to derive the "iterated Wiener filter" for the purpose ofdenoising images.