The Factored Frontier (FF) algorithm is a simple approximate inferencealgorithm for Dynamic Bayesian Networks (DBNs). It is very similar tothe fully factorized version of the Boyen-Koller (BK) algorithm, butinstead of doing an exact update at every step followed bymarginalisation (projection), it always works with factoreddistributions. Hence it can be applied to models for which the exactupdate step is intractable. We show that FF is equivalent to (oneiteration of) loopy belief propagation (LBP) on the original DBN, andthat BK is equivalent (to one iteration of) LBP on a DBN where wecluster some of the nodes. We then show empirically that byiterating, LBP can improve on the accuracy of both FF and BK. Wecompare these algorithms on two real-world DBNs: the first is a modelof a water treatment plant, and the second is a coupled HMM, used tomodel freeway traffic.

Simple Gaussian Mixture Models (GMMs) learned from pixels of natural image patches have been recently shown to be surprisingly strong performers in modeling the statistics of natural images. Here we provide an in depth analysis of this simple yet rich model. We show that such a GMM model is able to compete with even the most successful models of natural images in log likelihood scores, denoising performance and sample quality. We provide an analysis of what such a model learns from natural images as a function of number of mixture components - including covariance structure, contrast variation and intricate structures such as textures, boundaries and more. Finally, we show that the salient properties of the GMM learned from natural images can be derived from a simplified Dead Leaves model which explicitly models occlusion, explaining its surprising success relative to other models. 1 GMMs and natural image statistics models Many models for the statistics of natural image patches have been suggested in recent years.

Although human object recognition is supposedly robust to viewpoint, much research on human perception indicates that there is a preferred or “canonical” view of objects. This phenomenon was discovered more than 30 years ago but the canonical view of only a small number of categories has been validated experimentally. Moreover, the explanation for why humans prefer the canonical view over other views remains elusive. In this paper we ask: Can we use Internet image collections to learn more about canonical views? We start by manually finding the most common view in the results returned by Internet search engines when queried with the objects used in psychophysical experiments. Our results clearly show that the most likely view in the search engine corresponds to the same view preferred by human subjects in experiments. We also present a simple method to find the most likely view in an image collection and apply it to hundreds of categories. Using the new data we have collected we present strong evidence against the two most prominent formal theories of canonical views and provide novel constraints for new theories.

Finding the most probable assignment (MAP) in a general graphical model is known to be NP hard but good approximations have been attained with max-product belief propagation (BP) and its variants. In particular, it is known that using BP on a single-cycle graph or tree reweighted BP on an arbitrary graph will give the MAP solution if the beliefs have no ties. In this paper we extend the setting under which BP can be used to provably extract the MAP. We define Convex BP as BP algorithms based on a convex free energy approximation and show that this class includes ordinary BP with single-cycle, tree reweighted BP and many other BP variants. We show that when there are no ties, fixed-points of convex max-product BP will provably give the MAP solution. We also show that convex sum-product BP at sufficiently small temperatures can be used to solve linear programs that arise from relaxing the MAP problem. Finally, we derive a novel condition that allows us to derive the MAP solution even if some of the convex BP beliefs have ties. In experiments, we show that our theorems allow us to find the MAP in many real-world instances of graphical models where exact inference using junction-tree is impossible.

Latent topic models have been successfully applied as an unsupervised topic discovery technique in large document collections. With the proliferation of hypertext document collection such as the Internet, there has also been great interest in extending these approaches to hypertext [6, 9]. These approaches typically model links in an analogous fashion to how they model words - the document-link co-occurrence matrix is modeled in the same way that the document-word co-occurrence matrix is modeled in standard topic models. In this paper we present a probabilistic generative model for hypertext document collections that explicitly models the generation of links. Specifically, links from a word w to a document d depend directly on how frequent the topic of w is in d, in addition to the in-degree of d. We show how to perform EM learning on this model efficiently. By not modeling links as analogous to words, we end up using far fewer free parameters and obtain better link prediction results.

Message-passing algorithms have emerged as powerful techniques for approximate inference in graphical models. When these algorithms converge, they can be shown to find local (or sometimes even global) optima of variational formulations to the inference problem. But many of the most popular algorithms are not guaranteed to converge. This has lead to recent interest in convergent message-passing algorithms. In this paper, we present a unified view of convergent message-passing algorithms. We present a simple derivation of an abstract algorithm, tree-consistency bound optimization (TCBO) that is provably convergent in both its sum and max product forms. We then show that many of the existing convergent algorithms are instances of our TCBO algorithm, and obtain novel convergent algorithms "for free" by exchanging maximizations and summations in existing algorithms. In particular, we show that Wainwright's non-convergent sum-product algorithm for tree based variational bounds, is actually convergent with the right update order for the case where trees are monotonic chains.

We propose a new model for natural image statistics. Instead of minimizing dependency between components of natural images, we maximize a simple form of dependency in the form of tree-dependency. By learning filters and tree structures which are best suited for natural images we observe that the resulting filters are edge filters, similar to the famous ICA on natural images results. Calculating the likelihood of the model requires estimating the squared output of pairs of filters connected in the tree. We observe that after learning, these pairs of filters are predominantly of similar orientations but different phases, so their joint energy resembles models of complex cells.

With the advent of the Internet it is now possible to collect hundreds of millions of images. These images come with varying degrees of label information. ``Clean labels can be manually obtained on a small fraction, ``noisy labels may be extracted automatically from surrounding text, while for most images there are no labels at all. Semi-supervised learning is a principled framework for combining these different label sources. However, it scales polynomially with the number of images, making it impractical for use on gigantic collections with hundreds of millions of images and thousands of classes. In this paper we show how to utilize recent results in machine learning to obtain highly efficient approximations for semi-supervised learning that are linear in the number of images.  Specifically, we use the convergence of the eigenvectors of the normalized graph Laplacian to eigenfunctions of weighted Laplace-Beltrami operators. We combine this with a label sharing framework obtained from Wordnet to propagate label information to classes lacking manual annotations. Our algorithm enables us to apply semi-supervised learning to a database of 80 million images with 74 thousand classes.

Semantic hashing seeks compact binary codes of datapoints so that the Hamming distance between codewords correlates with semantic similarity. Hinton et al. used a clever implementation of autoencoders to find such codes. In this paper, we show that the problem of finding a best code for a given dataset is closely related to the problem of graph partitioning and can be shown to be NP hard. By relaxing the original problem, we obtain a spectral method whose solutions are simply a subset of thresh- olded eigenvectors of the graph Laplacian. By utilizing recent results on convergence of graph Laplacian eigenvectors to the Laplace-Beltrami eigen- functions of manifolds, we show how to efficiently calculate the code of a novel datapoint. Taken together, both learning the code and applying it to a novel point are extremely simple. Our experiments show that our codes significantly outperform the state-of-the art.

Sparse PCA seeks approximate sparse "eigenvectors" whose projections capture the maximal variance of data. As a cardinality-constrained and non-convex optimization problem, it is NPhard and is encountered in a wide range of applied fields, from bio-informatics to finance. Recent progress has focused mainly on continuous approximation and convex relaxation of the hard cardinality constraint. In contrast, we consider an alternative discrete spectral formulation based on variational eigenvalue bounds and provide an effective greedy strategy as well as provably optimal solutions using branch-and-bound search. Moreover, the exact methodology used reveals a simple renormalization step that improves approximate solutions obtained by any continuous method. The resulting performance gain of discrete algorithms is demonstrated on real-world benchmark data and in extensive Monte Carlo evaluation trials.