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

Reinforcement learning, one of the most active research areas in artificial intelligence, is a computational approach to learning whereby an agent tries to maximize the total amount of reward it receives when interacting with a complex, uncertain environment. In Reinforcement Learning, Richard Sutton and Andrew Barto provide a clear and simple account of the key ideas and algorithms of reinforcement learning. Their discussion ranges from the history of the field's intellectual foundations to the most recent developments and applications. The only necessary mathematical background is familiarity with elementary concepts of probability. The book is divided into three parts. Part I defines the reinforcement learning problem in terms of Markov decision processes. Part II provides basic solution methods: dynamic programming, Monte Carlo methods, and temporal-difference learning. Part III presents a unified view of the solution methods and incorporates artificial neural networks, eligibility traces, and planning; the two final chapters present case studies and consider the future of reinforcement learning.

/pdf/reinforcement-learning-an-introduction-rzxgej9p17.pdf

Reinforcement Learning: An Introduction

There is, I think, something ethereal about i —the square root of minus one. I remember first hearing about it at school. It seemed an odd beast at that time—an intruder hovering on the edge of reality.

Usually familiarity dulls this sense of the bizarre, but in the case of i it was the reverse: over the years the sense of its surreal nature intensified. It seemed that it was impossible to write mathematics that described the real world in …

I and i

The architecture and learning procedure underlying ANFIS (adaptive-network-based fuzzy inference system) is presented, which is a fuzzy inference system implemented in the framework of adaptive networks. By using a hybrid learning procedure, the proposed ANFIS can construct an input-output mapping based on both human knowledge (in the form of fuzzy if-then rules) and stipulated input-output data pairs. In the simulation, the ANFIS architecture is employed to model nonlinear functions, identify nonlinear components on-line in a control system, and predict a chaotic time series, all yielding remarkable results. Comparisons with artificial neural networks and earlier work on fuzzy modeling are listed and discussed. Other extensions of the proposed ANFIS and promising applications to automatic control and signal processing are also suggested. >

ANFIS: adaptive-network-based fuzzy inference system

https://www2.econ.iastate.edu/tesfatsi/DeepLearningInNeuralNetworksOverview.JSchmidhuber2015.pdf

Deep learning in neural networks

: In the paper, the direct method of Lyapunov is employed to outline a procedure for the identification of a discrete multivariable system from the measurements on the inputs and the state variables of the system It is assumed that all the state variables of the plant are accessible and that there is no measurement noise The procedure results in the convergence of every parameter of a model to the corresponding one of the system as the number of stages k nears infinity The overall process is also shown to be asymptotically stable (Author)

An Identification Procedure for Discrete Linear Multivariable Systems.

: A comparison is made of various methods presented in the literature for treating Linear systems with time-varying parameters. These variations are considered to arise both from small-amplitude signals of arbitrary frequency and also from signals which are much slower than the dominant signals of the system. Two time-domain approaches and one frequency-domain approach are shown to be equivalent, resulting in identical truncated series solutions, expressed in terms of a single time variable. The method is extended to the case where the slow variations are expressed as functions of a second-time-variable. The result is shown to be equivalent to a two time-scale approach described in the literature. A general unified approach is suggested encompassing all the above methods, whether for single- or double-time-scale series solutions. (Author)

On two time-scales in linear systems with special reference to adaptive control

https://ieeexplore.ieee.org/iel5/9/24101/01098919.pdf

Gradient generation using a single perturbation signal

: The main objective of this brief report is to analyze the asymptotic behavior of three error models which arise in the identification and control of hybrid adaptive systems. Using methods developed in 1 and 3 this analysis can be applied directly to establish the global stability of hybrid model reference adaptive systems. These results as well as the design considerations which arise in such problems will be contained in a forthcoming paper 8.

Error Models for Stable Hybrid Adaptive Control

A procedure based on multiple models, switching, and tuning for adaptively controlling dynamical systems in time-varying environments was introduced in the early 1990s. During the past decade, this procedure has evolved into a general methodology for adaptive control. Some of the recent advances in this new area are discussed in the paper.

Kumpati S. Narendra

Papers

An Identification Procedure for Discrete Linear Multivariable Systems.

On two time-scales in linear systems with special reference to adaptive control

Gradient generation using a single perturbation signal

Error Models for Stable Hybrid Adaptive Control

Adaptive Control Using Multiple Models: A Methodology