Deep learning allows computational models that are composed of multiple processing layers to learn representations of data with multiple levels of abstraction. These methods have dramatically improved the state-of-the-art in speech recognition, visual object recognition, object detection and many other domains such as drug discovery and genomics. Deep learning discovers intricate structure in large data sets by using the backpropagation algorithm to indicate how a machine should change its internal parameters that are used to compute the representation in each layer from the representation in the previous layer. Deep convolutional nets have brought about breakthroughs in processing images, video, speech and audio, whereas recurrent nets have shone light on sequential data such as text and speech.

Deep learning

Multilayer neural networks trained with the back-propagation algorithm constitute the best example of a successful gradient based learning technique. Given an appropriate network architecture, gradient-based learning algorithms can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters, with minimal preprocessing. This paper reviews various methods applied to handwritten character recognition and compares them on a standard handwritten digit recognition task. Convolutional neural networks, which are specifically designed to deal with the variability of 2D shapes, are shown to outperform all other techniques. Real-life document recognition systems are composed of multiple modules including field extraction, segmentation recognition, and language modeling. A new learning paradigm, called graph transformer networks (GTN), allows such multimodule systems to be trained globally using gradient-based methods so as to minimize an overall performance measure. Two systems for online handwriting recognition are described. Experiments demonstrate the advantage of global training, and the flexibility of graph transformer networks. A graph transformer network for reading a bank cheque is also described. It uses convolutional neural network character recognizers combined with global training techniques to provide record accuracy on business and personal cheques. It is deployed commercially and reads several million cheques per day.

Gradient-based learning applied to document recognition

We propose a deep convolutional neural network architecture codenamed Inception that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. By a carefully crafted design, we increased the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the Hebbian principle and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC14 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.

/pdf/going-deeper-with-convolutions-1yobw2o2ds.pdf

Going deeper with convolutions

Setting of the learning problem consistency of learning processes bounds on the rate of convergence of learning processes controlling the generalization ability of learning processes constructing learning algorithms what is important in learning theory?.

The Nature of Statistical Learning Theory

We propose a new framework for estimating generative models via an adversarial process, in which we simultaneously train two models: a generative model G that captures the data distribution, and a discriminative model D that estimates the probability that a sample came from the training data rather than G. The training procedure for G is to maximize the probability of D making a mistake. This framework corresponds to a minimax two-player game. In the space of arbitrary functions G and D, a unique solution exists, with G recovering the training data distribution and D equal to ½ everywhere. In the case where G and D are defined by multilayer perceptrons, the entire system can be trained with backpropagation. There is no need for any Markov chains or unrolled approximate inference networks during either training or generation of samples. Experiments demonstrate the potential of the framework through qualitative and quantitative evaluation of the generated samples.

/pdf/generative-adversarial-nets-1ofhan3sbf.pdf

Generative Adversarial Nets

A method and apparatus based on transposition to speed up learning computations on sparse data are disclosed. For example, the method receives an support vector comprising at least one feature represented by one non-zero entry. The method then identifies at least one column within a matrix with non-zero entries, wherein the at least one column is identified in accordance with the at least one feature of the support vector. The method then performs kernel computations using successive list merging on the at least one identified column of the matrix and the support vector to derive a result vector, wherein the result vector is used in a data learning function.

Method and apparatus for providing fast kernel learning on sparse data

Task 1 of the 2006 KDD Challenge Cup required classification of pulmonary embolisms (PEs) using variables derived from computed tomography angiography. We present our approach to the challenge and justification for our choices. We used boosted trees to perform the main classification task, but modified the algorithm to address idiosyncrasies of the scoring criteria. The two main modifications were: 1) changing the dependent variable in the training set to account for multiple PEs per patient, and 2) incorporating neighborhood information through augmentation of the set of predictor variables. Both of these resulted in measurable predictive improvement. In addition, we discuss a statistically based method for setting the classification threshold.

Modifying boosted trees to improve performance on task 1 of the 2006 KDD challenge cup

This paper introduces a methodology for speech data mining along with the tools that the methodology requires. We show how they increase the productivity of the analyst who seeks relationships among the contents of multiple utterances and ultimately must link some newly discovered context into testable hypotheses about new information. While in its simplest form, one can extend text data mining to speech data mining by using text tools on the output of a speech recognizer, we have found that it is not optimal. We show how data mining techniques that are typically applied to text should be modified to enable an analyst to do effective semantic data mining on a large collection of short speech utterances.

Patrick Haffner

Papers

Method and apparatus for providing fast kernel learning on sparse data

Modifying boosted trees to improve performance on task 1 of the 2006 KDD challenge cup

Semantic data mining of short utterances

Discriminant learning with minimum memory loss for improved non-vocabulary rejection.

Connectionist segmental post-processing of the n-best solutions in isolated and connected word recognition task.