No abstract

The fully-connected layers of deep convolutional neural networks typically contain over 90% of the network parameters. Reducing the number of parameters while preserving predictive performance is critically important for training big models in distributed systems and for deployment in embedded devices.In this paper, we introduce a novel Adaptive Fastfood transform to reparameterize the matrix-vector multiplication of fully connected layers. Reparameterizing a fully connected layer with d inputs and n outputs with the Adaptive Fastfood transform reduces the storage and computational costs costs from O(nd) to O(n) and O(n log d) respectively. Using the Adaptive Fastfood transform in convolutional networks results in what we call a deep fried convnet. These convnets are end-to-end trainable, and enable us to attain substantial reductions in the number of parameters without affecting prediction accuracy on the MNIST and ImageNet datasets.

In many classification problems labels are relatively scarce. One context in which this occurs is where we have labels for groups of instances but not for the instances themselves, as in multi-instance learning. Past work on this problem has typically focused on learning classifiers to make predictions at the group level. In this paper we focus on the problem of learning classifiers to make predictions at the instance level. To achieve this we propose a new objective function that encourages smoothness of inferred instance-level labels based on instance-level similarity, while at the same time respecting group-level label constraints. We apply this approach to the problem of predicting labels for sentences given labels for reviews, using a convolutional neural network to infer sentence similarity. The approach is evaluated using three large review data sets from IMDB, Yelp, and Amazon, and we demonstrate the proposed approach is both accurate and scalable compared to various alternatives.

We discuss an attentional model for simultaneous object tracking and recognition that is driven by gaze data. Motivated by theories of perception, the model consists of two interacting pathways, identity and control, intended to mirror the what and where pathways in neuroscience models. The identity pathway models object appearance and performs classification using deep (factored)-Restricted Boltzmann Machines. At each point in time the observations consist of foveated images, with decaying resolution toward the periphery of the gaze. The control pathway models the location, orientation, scale and speed of the attended object. The posterior distribution of these states is estimated with particle filtering. Deeper in the control pathway, we encounter an attentional mechanism that learns to select gazes so as to minimize tracking uncertainty. Unlike in our previous work, we introduce gaze selection strategies which operate in the presence of partial information and on a continuous action space. We show that a straightforward extension of the existing approach to the partial information setting results in poor performance, and we propose an alternative method based on modeling the reward surface as a Gaussian Process. This approach gives good performance in the presence of partial information and allows us to expand the action space from a small, discrete set of fixation points to a continuous domain.

This work presents a scalable solution to open-vocabulary visual speech recognition. To achieve this, we constructed the largest existing visual speech recognition dataset, consisting of pairs of text and video clips of faces speaking (3,886 hours of video). In tandem, we designed and trained an integrated lipreading system, consisting of a video processing pipeline that maps raw video to stable videos of lips and sequences of phonemes, a scalable deep neural network that maps the lip videos to sequences of phoneme distributions, and a production-level speech decoder that outputs sequences of words. The proposed system achieves a word error rate (WER) of 40.9% as measured on a held-out set. In comparison, professional lipreaders achieve either 86.4% or 92.9% WER on the same dataset when having access to additional types of contextual information. Our approach significantly improves on other lipreading approaches, including variants of LipNet and of Watch, Attend, and Spell (WAS), which are only capable of 89.8% and 76.8% WER respectively. * These authors contributed equally to this work.

Abstract. In this paper we give a systematic analysis of the relationship between imbalance and overlap as factors influencing classifier performance. We demonstrate that these two factors have interdependent effects and that we cannot form a full understanding of their effects by considering them only in isolation. Although the imbalance problem can be considered a symptom of the small disjuncts problem which is solved by using larger training sets, the overlap problem is of a fundamentally different character and the performance of learned classifiers can actually be made worse by using more training data when overlap is present. We also examine the effects of overlap and imbalance on the complexity of the learned model and demonstrate that overlap is a far more serious factor than imbalance in this respect.

No abstract

No abstract