A concise tutorial description of the cellular neural network (CNN) paradigm is given, along with a precise taxonomy. The CNN is defined, and the canonical equations are described. The importance of many independent input signal arrays, adaptive templates, and the multilayer capability is emphasized and motivated by examples. It is shown how simply a wave-type partial differential equation can be generated. >

The CNN paradigm

https://www.cse.iitk.ac.in/users/isaha/Publications/Journals/NC10.pdf/

Artificial neural networks in hardware: A survey of two decades of progress

Neuromorphic computing has come to refer to a variety of brain-inspired computers, devices, and models that contrast the pervasive von Neumann computer architecture This biologically inspired approach has created highly connected synthetic neurons and synapses that can be used to model neuroscience theories as well as solve challenging machine learning problems The promise of the technology is to create a brain-like ability to learn and adapt, but the technical challenges are significant, starting with an accurate neuroscience model of how the brain works, to finding materials and engineering breakthroughs to build devices to support these models, to creating a programming framework so the systems can learn, to creating applications with brain-like capabilities In this work, we provide a comprehensive survey of the research and motivations for neuromorphic computing over its history We begin with a 35-year review of the motivations and drivers of neuromorphic computing, then look at the major research areas of the field, which we define as neuro-inspired models, algorithms and learning approaches, hardware and devices, supporting systems, and finally applications We conclude with a broad discussion on the major research topics that need to be addressed in the coming years to see the promise of neuromorphic computing fulfilled The goals of this work are to provide an exhaustive review of the research conducted in neuromorphic computing since the inception of the term, and to motivate further work by illuminating gaps in the field where new research is needed

A Survey of Neuromorphic Computing and Neural Networks in Hardware.

The paper introduces a general class of neural networks where the neuron activations are modeled by discontinuous functions. The neural networks have an additive interconnecting structure and they include as particular cases the Hopfield neural networks (HNNs), and the standard cellular neural networks (CNNs), in the limiting situation where the HNNs and CNNs possess neurons with infinite gain. Conditions are derived which ensure the existence of a unique equilibrium point, and a unique output equilibrium point, which are globally attractive for the state and the output trajectories of the neural network, respectively. These conditions, which are applicable to general nonsymmetric neural networks, are based on the concept of Lyapunov diagonally-stable neuron interconnection matrices, and they can be thought of as a generalization to the discontinuous case of previous results established for neural networks possessing smooth neuron activations. Moreover, by suitably exploiting the presence of sliding modes, entirely new conditions are obtained which ensure global convergence in finite time, where the convergence time can be easily estimated on the basis of the relevant neural-network parameters. The analysis in the paper employs results from the theory of differential equations with discontinuous right-hand side as introduced by Filippov. In particular, global convergence is addressed by using a Lyapunov-like approach based on the concept of monotone trajectories of a differential inclusion.

Global convergence of neural networks with discontinuous neuron activations

This paper presents a comprehensive review of the industrial applications of artificial neural networks (ANNs), in the last 12 years. Common questions that arise to practitioners and control engineers while deciding how to use NNs for specific industrial tasks are answered. Workable issues regarding implementation details, training and performance evaluation of such algorithms are also discussed, based on a judiciously chronological organization of topologies and training methods effectively used in the past years. The most popular ANN topologies and training methods are listed and briefly discussed, as a reference to the application engineer. Finally, ANN industrial applications are grouped and tabulated by their main functions and what they actually performed on the referenced papers. The authors prepared this paper bearing in mind that an organized and normalized review would be suitable to help industrial managing and operational personnel decide which kind of ANN topology and training method would be adequate for their specific problems.

A comprehensive review for industrial applicability of artificial neural networks

A network structure called a discrete-time cellular neural network is introduced. It is derived from cellular neural networks and feedback threshold networks. the architecture is discussed and its advantages are presented. Convergence is proved for a large class of templates and applications are given for the following image-processing tasks: linear thresholding, connected component detection, hole filling, concentric contouring, increasing and decreasing objects step by step, searching for objects with minimal distance, and oscillation.

Discrete-time cellular neural networks

An analog circuit structure for the realization of discrete-time cellular neural networks (DTCNNs) is introduced. The computation is done by a balanced clocked circuit based on the idea of conductance multipliers and operational transconductance amplifiers. The circuit is proposed for a one-neighborhood on a hexagonal grid, but can also be modified to larger neighborhoods and/or other grid topologies. A layout was designed for a standard CMOS process, and the corresponding HSPICE simulation results are given. A test chip containing 16 cells was fabricated, and measurements of the transfer characteristics are provided. The functional behavior is demonstrated for a simple example. >

An analog implementation of discrete-time cellular neural networks

A generalized architecture for discrete-time cellular neural networks (DTCNNs) is provided. It allows multiple layers of different architecture, which can be combined in several interconnection modes. Three elementary building blocks are introduced. Another important extension is the use of time-variant templates. They allow the definition of cyclic templates, where the coefficients are changed every iteration step and a set of templates is applied periodically. The definition of convergence has been adopted for this network structure, and important classes of templates are proved to be convergent. Examples are given for the following image processing tasks: rectangular hull extraction, skeletonization, and halftoning. >

Multiple layer discrete-time cellular neural networks using time-variant templates

The paper describes an analog current mode realization of Discrete-Time Cellular Neural Networks (DTCNNs) with high cell density, which have local analog and local logic memory. Hence, some important parts of the CNN Universal Machine concept are implemented. The computation speed can be adjusted simply to the application by changing the clock rate. The circuit components are described in detail and SPICE level 2 simulation results are given for the ORBIT 2.0 /spl mu/m process. A layout has been designed for a chip with 12 by 12 cells on a square grid realizing a one-neighborhood with 9 feedback and 9 control coefficients. The cell size is 619 /spl mu/m by 425 /spl mu/m and the simulated speed is between 1 MHz and 10 MHz depending on the minimum value of the state current. For the latter this leads to a simulated performance of 25.9 10/sup 9/ XPS for a single chip operation with an effective area of 0.379 cm/sup 2/ and a worst case power consumption of 0.86 W. Another important feature of the chip is its capability for a spatial cascaded connection. >

A current-mode DTCNN universal chip

The paper gives two examples, where an analog input image is preprocessed by a sequence of templates, i.e. by analogic CNN algorithms running on the CNN Universal Machine. The examples are: the extraction of horizontal screws with arbitrary length and the classification of screws according to their size. >

Hubert Harrer

Papers

Discrete-time cellular neural networks

An analog implementation of discrete-time cellular neural networks

Multiple layer discrete-time cellular neural networks using time-variant templates

A current-mode DTCNN universal chip

Some examples of preprocessing analog images with discrete-time cellular neural networks