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

The aim of this paper is to communicate to a broader audience in systems theory the relevance of dynamical phenomena that are described by fractional order derivatives. It is organized with four principal objectives in mind. First, it attempts to introduce the reader to the rich mathematical theory behind the definition of the term "fractional order derivative." Following this, it highlights the diversity of real-life problems in engineering, economics and the life sciences to which researchers have successfully applied models containing fractional order derivatives. Then, as an essential step for the consideration of such models in systems theory, it summarizes the state of research in the representation and analysis of dynamical systems with fractional order derivatives (also referred to as fractional systems). The paper concludes with some comments aimed at the development of a more general theory of fractional systems.

Fractional Order Derivatives in Systems Theory.

Necessary and sufficient conditions are derived in this paper for the existence of a diagonal Lyapunov function for a class of switched systems. The switching is carried out intermittently between the open loop and the closed loop of a feedback system, a procedure encountered frequently in many interconnected systems at the present time. The open loop system considered in the paper is described by a stable Metzler matrix, which is known to be diagonally stable. The closed loop system is stable, but does not retain the Metzler property. The existence of a common diagonal Lyapunov function assures system stability for arbitrary switching. This decouples stability and performance issues in the switched system, and permits switching to be carried out entirely for improving performance. Towards the end of the paper, it is also shown that some recent results in the control literature can be derived as special cases of the principal results derived in the paper.

On the diagonal stability of a class of almost positive switched systems

This paper attempts a unified presentation of general schemes using Lyapunov's direct method for (i) the adaptive control and (ii) the identification of multivariable systems. The first part deals with cases where all the state variables of the plant are accessible. In the second part, an adaptive observer is synthesized to identify the plant parameters and to simultaneously estimate the plant state vector, when only the output of the system is accessible.

Identification and adaptation using Lyaponov's direct method

https://ieeexplore.ieee.org/iel5/9/24104/01099043.pdf

Kumpati S. Narendra

Papers

Fractional Order Derivatives in Systems Theory.

On the diagonal stability of a class of almost positive switched systems

Identification and adaptation using Lyaponov's direct method

Stability of the damped Mathieu equation

Robust Adaptive Control Using Reduced Order Models