Richard S. Zemel

The helmholtz machine

Discovering the structure inherent in a set of patterns is a fundamental aim of statistical inference or learning. One fruitful approach is to build a parameterized stochastic generative model, independent draws from which are likely to produce the patterns. For all but the simplest generative models, each pattern can be generated in exponentially many ways. It is thus intractable to adjust the parameters to maximize the probability of the observed patterns. We describe a way of finessing this combinatorial explosion by maximizing an easily computed lower bound on the probability of the observations. Our method can be viewed as a form of hierarchical self-supervised learning that may relate to the function of bottom-up and top-down cortical processing pathways.

Multiscale conditional random fields for image labeling

We propose an approach to include contextual features for labeling images, in which each pixel is assigned to one of a finite set of labels. The features are incorporated into a probabilistic framework, which combines the outputs of several components. Components differ in the information they encode. Some focus on the image-label mapping, while others focus solely on patterns within the label field. Components also differ in their scale, as some focus on fine-resolution patterns while others on coarser, more global structure. A supervised version of the contrastive divergence algorithm is applied to learn these features from labeled image data. We demonstrate performance on two real-world image databases and compare it to a classifier and a Markov random field.

Probabilistic Interpretation of Population Codes

We present a general encoding-decoding framework for interpreting the activity of a population of units. A standard population code interpretation method, the Poisson model, starts from a description as to how a single value of an underlying quantity can generate the activities of each unit in the population. In casting it in the encoding-decoding framework, we find that this model is too restrictive to describe fully the activities of units in population codes in higher processing areas, such as the medial temporal area. Under a more powerful model, the population activity can convey information not only about a single value of some quantity but also about its whole distribution, including its variance, and perhaps even the certainty the system has in the actual presence in the world of the entity generating this quantity. We propose a novel method for forming such probabilistic interpretations of population codes and compare it to the existing method.

Object-Based Attention and Occlusion Evidence From Normal Participants and a Computational Model

One way of perceptually organizing a complex visual scene is to attend selectively to information in a particular physical location. Another way of reducing the complexity in the input is to attend selectively to an individual object in the scene and to process its elements preferentially. This latter, object-based attention process was examined, and the predicted superiority for reporting features from 1 relative to 2 objects was replicated in a series of experiments. This object-based process was robust even under conditions of occlusion, although there were some boundary conditions on its operation. Finally, an account of the data is provided via simulations of the findings in a computational model. The claim is that object-based attention arises from a mechanisms that groups together those features based on internal representations developed over perceptual experience and then preferentially gates these features for later, selective processing.

Learning and incorporating top-down cues in image segmentation

Bottom-up approaches, which rely mainly on continuity principles, are often insufficient to form accurate segments in natural images. In order to improve performance, recent methods have begun to incorporate top-down cues, or object information, into segmentation. In this paper, we propose an approach to utilizing category-based information in segmentation, through a formulation as an image labelling problem. Our approach exploits bottom-up image cues to create an over-segmented representation of an image. The segments are then merged by assigning labels that correspond to the object category. The model is trained on a database of images, and is designed to be modular: it learns a number of image contexts, which simplify training and extend the range of object classes and image database size that the system can handle. The learning method estimates model parameters by maximizing a lower bound of the data likelihood. We examine performance on three real-world image databases, and compare our system to a standard classifier and other conditional random field approaches, as well as a bottom-up segmentation method.

Aligning Books and Movies: Towards Story-Like Visual Explanations by Watching Movies and Reading Books

Books are a rich source of both fine-grained information, how a character, an object or a scene looks like, as well as high-level semantics, what someone is thinking, feeling and how these states evolve through a story. This paper aims to align books to their movie releases in order to provide rich descriptive explanations for visual content that go semantically far beyond the captions available in the current datasets. To align movies and books we propose a neural sentence embedding that is trained in an unsupervised way from a large corpus of books, as well as a video-text neural embedding for computing similarities between movie clips and sentences in the book. We propose a context-aware CNN to combine information from multiple sources. We demonstrate good quantitative performance for movie/book alignment and show several qualitative examples that showcase the diversity of tasks our model can be used for.

Collaborative prediction and ranking with non-random missing data

A fundamental aspect of rating-based recommender systems is the observation process, the process by which users choose the items they rate. Nearly all research on collaborative filtering and recommender systems is founded on the assumption that missing ratings are missing at random. The statistical theory of missing data shows that incorrect assumptions about missing data can lead to biased parameter estimation and prediction. In a recent study, we demonstrated strong evidence for violations of the missing at random condition in a real recommender system. In this paper we present the first study of the effect of non-random missing data on collaborative ranking, and extend our previous results regarding the impact of non-random missing data on collaborative prediction.

A Model for Encoding Multiple Object Motions and Self-Motion in Area MST of Primate Visual Cortex

Many cells in the dorsal part of the medial superior temporal (MST) region of visual cortex respond selectively to specific combinations of expansion/contraction, translation, and rotation motions. Previous investigators have suggested that these cells may respond selectively to the flow fields generated by self-motion of an observer. These patterns can also be generated by the relative motion between an observer and a particular object. We explored a neurally constrained model based on the hypothesis that neurons in MST partially segment the motion fields generated by several independently moving objects. Inputs to the model were generated from sequences of ray-traced images that simulated realistic motion situations, combining observer motion, eye movements, and independent object motions. The input representation was based on the response properties of neurons in the middle temporal area (MT), which provides the primary input to area MST. After applying an unsupervised optimization technique, the units became tuned to patterns signaling coherent motion, matching many of the known properties of MST cells. The results of this model are consistent with recent studies indicating that MST cells primarily encode information concerning the relative three-dimensional motion between objects and the observer.

Experience-dependent perceptual grouping and object-based attention

Earlier studies have shown that attention can be directed to objects, defined on the basis of generic grouping principles, highly familiar shapes, or task instructions, rather than to contiguous regions of the visual field. The 4 experiments presented in this article extend these findings, showing that object attention benefits—shorter reaction times to features appearing on a single object—apply to recently viewed novel shapes. One experiment shows that object attention operates even when the visible fragments correspond to objects that violate standard completion heuristics. Other experiments show that experience-dependent object benefits can apply to fragments even without evidence of occlusion. These results attest to the flexible operation of the perceptual system, adapting as a function of experience.

BoltzRank: learning to maximize expected ranking gain

Ranking a set of retrieved documents according to their relevance to a query is a popular problem in information retrieval. Methods that learn ranking functions are difficult to optimize, as ranking performance is typically judged by metrics that are not smooth. In this paper we propose a new listwise approach to learning to rank. Our method creates a conditional probability distribution over rankings assigned to documents for a given query, which permits gradient ascent optimization of the expected value of some performance measure. The rank probabilities take the form of a Boltzmann distribution, based on an energy function that depends on a scoring function composed of individual and pairwise potentials. Including pairwise potentials is a novel contribution, allowing the model to encode regularities in the relative scores of documents; existing models assign scores at test time based only on individual documents, with no pairwise constraints between documents. Experimental results on the LETOR3.0 data set show that our method out-performs existing learning approaches to ranking.