In this work we investigate the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting. Our main contribution is a thorough evaluation of networks of increasing depth using an architecture with very small (3x3) convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers. These findings were the basis of our ImageNet Challenge 2014 submission, where our team secured the first and the second places in the localisation and classification tracks respectively. We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. We have made our two best-performing ConvNet models publicly available to facilitate further research on the use of deep visual representations in computer vision.

Very Deep Convolutional Networks for Large-Scale Image 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

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

Training Deep Neural Networks is complicated by the fact that the distribution of each layer's inputs changes during training, as the parameters of the previous layers change. This slows down the training by requiring lower learning rates and careful parameter initialization, and makes it notoriously hard to train models with saturating nonlinearities. We refer to this phenomenon as internal covariate shift, and address the problem by normalizing layer inputs. Our method draws its strength from making normalization a part of the model architecture and performing the normalization for each training mini-batch. Batch Normalization allows us to use much higher learning rates and be less careful about initialization, and in some cases eliminates the need for Dropout. Applied to a state-of-the-art image classification model, Batch Normalization achieves the same accuracy with 14 times fewer training steps, and beats the original model by a significant margin. Using an ensemble of batch-normalized networks, we improve upon the best published result on ImageNet classification: reaching 4.82% top-5 test error, exceeding the accuracy of human raters.

/pdf/batch-normalization-accelerating-deep-network-training-by-3dkqye6twt.pdf

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Abstract The past decade has seen tremendous progress in the field of artificial intelligence thanks to the resurgence of neural networks through deep learning. This has helped improve the ability for computers to see, hear, and understand the world around them, leading to dramatic advances in the application of AI to many fields of science and other areas of human endeavor. In this essay, I examine the reasons for this progress, including the confluence of progress in computing hardware designed to accelerate machine learning and the emergence of open-source software frameworks to dramatically expand the set of people who can use machine learning effectively. I also present a broad overview of some of the areas in which machine learning has been applied over the past decade. Finally, I sketch out some likely directions from which further progress in artificial intelligence will come.

/pdf/a-golden-decade-of-deep-learning-computing-systems-2ckbe8pc.pdf

A Golden Decade of Deep Learning: Computing Systems & Applications

Storing and Moving Data in a Distributed Storage System

In this talk I will describe the development of systems that underlie modern cloud computing systems. This development shares much of its motivation with the related fields of transaction processing systems and high performance computing, but because of scale, these systems tend to have more emphasis on fault tolerance using software techniques. Important developments in the development of modern cloud systems include very high performance distributed file system, such as the Google File System (Ghemawat et al., SOSP 2003), reliable computational frameworks such as MapReduce (Dean & Ghemawat, OSDI 2004) and Dryad (Isard et al., 2007), and large scale structured storage systems such as BigTable (Chang et al. 2006), Dynamo (DeCandia et al., 2007), and Spanner (Corbett et al., 2012). Scheduling computations can either be done using virtual machines (exemplified by VMWare's products), or as individual processes or containers. The development of public cloud platforms such as AWS, Microsoft Azure, and Google Cloud Platform, allow external developers to utilize these large-scale services to build new and interesting services and products, benefiting from the economies of scale of large datacenters and the ability to grow and shrink computing resources on demand across millions of customers.

The rise of cloud computing systems

A system generates a score for a document based on best seller list information relating to the document, circulation information relating to the document, publisher sales information relating to a publisher associated with the document, author sales information relating to an author associated with the document, user behavior information relating to access of the document by users, and/or a link-based score of a web document relating to the document.

Jeffrey Dean

Papers

A Golden Decade of Deep Learning: Computing Systems & Applications

Storing and Moving Data in a Distributed Storage System

The rise of cloud computing systems

Systems and methods for scoring documents

Translating terms using numeric representations