Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers—8× deeper than VGG nets [40] but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions1, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

/pdf/deep-residual-learning-for-image-recognition-b75jg61qrq.pdf

Deep Residual Learning for Image Recognition

Learning to store information over extended time intervals by recurrent backpropagation takes a very long time, mostly because of insufficient, decaying error backflow. We briefly review Hochreiter's (1991) analysis of this problem, then address it by introducing a novel, efficient, gradient based method called long short-term memory (LSTM). Truncating the gradient where this does not do harm, LSTM can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units. Multiplicative gate units learn to open and close access to the constant error flow. LSTM is local in space and time; its computational complexity per time step and weight is O. 1. Our experiments with artificial data involve local, distributed, real-valued, and noisy pattern representations. In comparisons with real-time recurrent learning, back propagation through time, recurrent cascade correlation, Elman nets, and neural sequence chunking, LSTM leads to many more successful runs, and learns much faster. LSTM also solves complex, artificial long-time-lag tasks that have never been solved by previous recurrent network algorithms.

Long short-term memory

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

Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers---8x deeper than VGG nets but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. 
The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

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

: This paper provides a brief overview of the connectionist or parallel distributed processing framework for modeling cognitive processes, and considers the application of the connectionist framework to problems of cognitive development. Several aspects of cognitive development might result from the process of learning as it occurs in multi-layer networks. This learning process has the characteristic that it reduces the discrepancy between expected and observed events. As it does this, representations develop on hidden units which dramatically change both the way in which the network represents the environment from which it learns and the expectations that the network generates about environmental events. The learning process exhibits relatively abrupt transitions corresponding to stage shifts in cognitive development. These points are illustrated using a network that learns to anticipate which side of a balance beam will go down, based on the number of weights on each side of the fulcrum and their distance from the fulcrum on each side of the beam. The network is trained in an environment in which weight more frequently governs which side will go down. It recapitulates the states of development seen in children, as well as the stage transitions, as it learns to represent weight and distance information. Keywords: Parallel processing; Data processing.

/pdf/parallel-distributed-processing-implications-for-cognition-2o5qmgcjgw.pdf

Parallel distributed processing: Implications for cognition and development.

The N400 component of the event-related brain potential has aroused much interest because it is thought to provide an online measure of meaning processing in the brain. However, the underlying process remains incompletely understood and actively debated. Here we present a computationally explicit account of this process and the emerging representation of sentence meaning. We simulate N400 amplitudes as the change induced by an incoming stimulus in an implicit and probabilistic representation of meaning captured by the hidden unit activation pattern in a neural network model of sentence comprehension, and we propose that the process underlying the N400 also drives implicit learning in the network. The model provides a unified account of 16 distinct findings from the N400 literature and connects human language comprehension with recent deep learning approaches to language processing.

/pdf/modelling-the-n400-brain-potential-as-change-in-a-4htin48hp2.pdf

Modelling the N400 brain potential as change in a probabilistic representation of meaning.

http://languagelog.ldc.upenn.edu/myl/ldc/papers/McClelland1991.pdf

Stochastic Interactive Processes and the Effect of Context on Perception

Deficits in irregular past-tense verb morphology associated with degraded semantic knowledge.

We describe a new learning procedure for networks that contain groups of nonlinear units arranged in a closed loop. The aim of the learning is to discover codes that allow the activity vectors in a "visible" group to be represented by activity vectors in a "hidden" group. One way to test whether a code is an accurate representation is to try to reconstruct the visible vector from the hidden vector. The difference between the original and the reconstructed visible vectors is called the reconstruction error, and the learning procedure aims to minimize this error. The learning procedure has two passes. On the first pass, the original visible vector is passed around the loop, and on the second pass an average of the original vector and the reconstructed vector is passed around the loop. The learning procedure changes each weight by an amount proportional to the product of the "presynaptic" activity and the difference in the post-synaptic activity on the two passes. This procedure is much simpler to implement than methods like back-propagation. Simulations in simple networks show that it usually converges rapidly on a good set of codes, and analysis shows that in certain restricted cases it performs gradient descent in the squared reconstruction error.

/pdf/learning-representations-by-recirculation-3w2gzagga0.pdf

James L. McClelland

Papers

Parallel distributed processing: Implications for cognition and development.

Modelling the N400 brain potential as change in a probabilistic representation of meaning.

Stochastic Interactive Processes and the Effect of Context on Perception

Deficits in irregular past-tense verb morphology associated with degraded semantic knowledge.

Learning Representations by Recirculation