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 this paper, necessary and sufficient conditions are derived for the existence of a common quadra-tic Lyapunov function for a finite number of stable second order linear time-invariant systems. Copyright © 2002 John Wiley & Sons, Ltd.

Necessary and sufficient conditions for the existence of a common quadratic Lyapunov function for a finite number of stable second order linear time-invariant systems

In this note, the problem of determining necessary and sufficient conditions for the existence of a common quadratic Lyapunov function for a pair of stable linear time-invariant systems whose system matrices A/sub 1/ and A/sub 2/ are in companion form is considered. It is shown that a necessary and sufficient condition for the existence of such a function is that the matrix product A/sub 1/A/sub 2/ does not have an eigenvalue that is real and negative. Examples are presented to illustrate the result.

/pdf/on-common-quadratic-lyapunov-functions-for-pairs-of-stable-3kathiqbte.pdf

On common quadratic Lyapunov functions for pairs of stable LTI systems whose system matrices are in companion form

Adaptive control using neural networks

Since 1990 a substantial amount of research has been reported in the literature concerning the identification and control of nonlinear dynamical systems using artificial neural networks. Various methods for tracking based on inverse control have been proposed, and constitute one of the main thrusts of this research effort. A significant part of this work has been heuristic in nature, and the conclusions drawn are generally justified using computer simulations. The general success of the simulation studies has also resulted in the increased use of artificial neural networks as controllers in industrial applications. As a result, there is a real need for a better understanding of the questions and problems that can arise in such contexts. This paper attempts to provide the theoretical foundations as well as insights that are essential for the efficient design of neural network controllers based on inverse control.

Issues in the application of neural networks for tracking based on inverse control

An adaptive scheme is synthesized to estimate the state vector and identify all the parameters of a Single-input single-output nth order linear system, from only the input—output data. The scheme is proved to be asymptotically stable in the whole, thus ensuring the convergence of the identification process.

Kumpati S. Narendra

Papers

Necessary and sufficient conditions for the existence of a common quadratic Lyapunov function for a finite number of stable second order linear time-invariant systems

On common quadratic Lyapunov functions for pairs of stable LTI systems whose system matrices are in companion form

Adaptive control using neural networks

Issues in the application of neural networks for tracking based on inverse control

Synthesis of an adaptive observer using Lyapunov's direct method