We trained a large, deep convolutional neural network to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes. On the test data, we achieved top-1 and top-5 error rates of 37.5% and 17.0% which is considerably better than the previous state-of-the-art. The neural network, which has 60 million parameters and 650,000 neurons, consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully-connected layers with a final 1000-way softmax. To make training faster, we used non-saturating neurons and a very efficient GPU implementation of the convolution operation. To reduce overriding in the fully-connected layers we employed a recently-developed regularization method called "dropout" that proved to be very effective. We also entered a variant of this model in the ILSVRC-2012 competition and achieved a winning top-5 test error rate of 15.3%, compared to 26.2% achieved by the second-best entry.

/pdf/imagenet-classification-with-deep-convolutional-neural-1190x9a6b9.pdf

ImageNet Classification with Deep Convolutional Neural Networks

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

LIBSVM is a library for Support Vector Machines (SVMs). We have been actively developing this package since the year 2000. The goal is to help users to easily apply SVM to their applications. LIBSVM has gained wide popularity in machine learning and many other areas. In this article, we present all implementation details of LIBSVM. Issues such as solving SVM optimization problems theoretical convergence multiclass classification probability estimates and parameter selection are discussed in detail.

/pdf/libsvm-a-library-for-support-vector-machines-2gdxasg9zv.pdf

LIBSVM: A library for support vector machines

Deep learning is a form of machine learning that enables computers to learn from experience and understand the world in terms of a hierarchy of concepts. Because the computer gathers knowledge from experience, there is no need for a human computer operator to formally specify all the knowledge that the computer needs. The hierarchy of concepts allows the computer to learn complicated concepts by building them out of simpler ones; a graph of these hierarchies would be many layers deep. This book introduces a broad range of topics in deep learning. The text offers mathematical and conceptual background, covering relevant concepts in linear algebra, probability theory and information theory, numerical computation, and machine learning. It describes deep learning techniques used by practitioners in industry, including deep feedforward networks, regularization, optimization algorithms, convolutional networks, sequence modeling, and practical methodology; and it surveys such applications as natural language processing, speech recognition, computer vision, online recommendation systems, bioinformatics, and videogames. Finally, the book offers research perspectives, covering such theoretical topics as linear factor models, autoencoders, representation learning, structured probabilistic models, Monte Carlo methods, the partition function, approximate inference, and deep generative models. Deep Learning can be used by undergraduate or graduate students planning careers in either industry or research, and by software engineers who want to begin using deep learning in their products or platforms. A website offers supplementary material for both readers and instructors.

Deep Learning

Deep neural nets with a large number of parameters are very powerful machine learning systems. However, overfitting is a serious problem in such networks. Large networks are also slow to use, making it difficult to deal with overfitting by combining the predictions of many different large neural nets at test time. Dropout is a technique for addressing this problem. The key idea is to randomly drop units (along with their connections) from the neural network during training. This prevents units from co-adapting too much. During training, dropout samples from an exponential number of different "thinned" networks. At test time, it is easy to approximate the effect of averaging the predictions of all these thinned networks by simply using a single unthinned network that has smaller weights. This significantly reduces overfitting and gives major improvements over other regularization methods. We show that dropout improves the performance of neural networks on supervised learning tasks in vision, speech recognition, document classification and computational biology, obtaining state-of-the-art results on many benchmark data sets.

/pdf/dropout-a-simple-way-to-prevent-neural-networks-from-4gtr92okhv.pdf

Dropout: a simple way to prevent neural networks from overfitting

This paper applies fast sparse multidimensional scaling (MDS) to a large graph of music similarity, with 267K vertices that represent artists, albums, and tracks; and 3.22M edges that represent similarity between those entities. Once vertices are assigned locations in a Euclidean space, the locations can be used to browse music and to generate playlists.

MDS on very large sparse graphs can be effectively performed by a family of algorithms called Rectangular Dijsktra (RD) MDS algorithms. These RD algorithms operate on a dense rectangular slice of the distance matrix, created by calling Dijsktra a constant number of times. Two RD algorithms are compared: Landmark MDS, which uses the Nystrom approximation to perform MDS; and a new algorithm called Fast Sparse Embedding, which uses FastMap. These algorithms compare favorably to Laplacian Eigenmaps, both in terms of speed and embedding quality.

/pdf/fast-embedding-of-sparse-music-similarity-graphs-53zgz085w0.pdf

Fast embedding of sparse music similarity graphs

Contextual advertising on web pages has become very popular recently and it poses its own set of unique text mining challenges. Often advertisers wish to either target (or avoid) some specific content on web pages which may appear only in a small part of the page. Learning for these targeting tasks is difficult since most training pages are multi-topic and need expensive human labeling at the sub-document level for accurate training. In this paper we investigate ways to learn for sub-document classification when only page level labels are available - these labels only indicate if the relevant content exists in the given page or not. We propose the application of multiple-instance learning to this task to improve the effectiveness of traditional methods. We apply sub-document classification to two different problems in contextual advertising. One is "sensitive content detection" where the advertiser wants to avoid content relating to war, violence, pornography, etc. even if they occur only in a small part of a page. The second problem involves opinion mining from review sites - the advertiser wants to detect and avoid negative opinion about their product when positive, negative and neutral sentiments co-exist on a page. In both these scenarios we present experimental results to show that our proposed system is able to get good block level labeling for free and improve the performance of traditional learning methods.

/pdf/learning-from-multi-topic-web-documents-for-contextual-idj9ykrv02.pdf

Learning from multi-topic web documents for contextual advertisement

Data sets involving multiple groups with shared characteristics frequently arise in practice. In this paper we extend hierarchical Dirichlet processes to model such data. Each group is assumed to be generated from a template mixture model with group level variability in both the mixing proportions and the component parameters. Variabilities in mixing proportions across groups are handled using hierarchical Dirichlet processes, also allowing for automatic determination of the number of components. In addition, each group is allowed to have its own component parameters coming from a prior described by a template mixture model. This group-level variability in the component parameters is handled using a random effects model. We present a Markov Chain Monte Carlo (MCMC) sampling algorithm to estimate model parameters and demonstrate the method by applying it to the problem of modeling spatial brain activation patterns across multiple images collected via functional magnetic resonance imaging (fMRI).

Hierarchical Dirichlet Processes with Random Effects

This paper describes the use of a convolutional neural network to perform address block location on machine-printed mail pieces. Locating the address block is a difficult object recognition problem because there is often a large amount of extraneous printing on a mail piece and because address blocks vary dramatically in size and shape.

We used a convolutional locator network with four outputs, each trained to find a different corner of the address block. A simple set of rules was used to generate ABL candidates from the network output. The system performs very well: when allowed five guesses, the network will tightly bound the address delivery information in 98.2% of the cases.

/pdf/postal-address-block-location-using-a-convolutional-locator-1p2fil22x5.pdf

Postal Address Block Location Using a Convolutional Locator Network

A “Music Mapper” automatically constructs a set coordinate vectors for use in inferring similarity between various pieces of music. In particular, given a music similarity graph expressed as links between various artists, albums, songs, etc., the Music Mapper applies a recursive embedding process to embed each of the graphs music entries into a multi-dimensional space. This recursive embedding process also embeds new music items added to the music similarity graph without reembedding existing entries so long a convergent embedding solution is achieved. Given this embedding, coordinate vectors are then computed for each of the embedded musical items. The similarity between any two musical items is then determined as either a function of the distance between the two corresponding vectors. In various embodiments, this similarity is then used in constructing music playlists given one or more random or user selected seed songs or in a statistical music clustering process.

John Platt

Papers

Fast embedding of sparse music similarity graphs

Learning from multi-topic web documents for contextual advertisement

Hierarchical Dirichlet Processes with Random Effects

Postal Address Block Location Using a Convolutional Locator Network

Client-based generation of music playlists from a server-provided subset of music similarity vectors