Novel Biomimetic Si Devices for Neuromorphic Computing Architecture.

SpringerLink Springer Engineering eBooks 2017 English+International

Incluye referencias bibliográficas e índices

Preface; Contents; Dr. Manan Suri; Hardware Spiking Artificial Neurons, Their Response Function, and Noises; 1 Introduction; 1.1 Biological Neurons; 1.2 Neuronal Response Function; 1.3 Neuronal Noises; 1.4 Artificial Neuron Models; 2 Hardware Spiking Neurons; 2.1 Silicon Neurons; 2.2 Emerging Spiking Neurons; 3 Summary and Outlook; References; Synaptic Plasticity with Memristive Nanodevices; 1 Introduction; 2 Neuromorphic Systems: Basic Processing and Data Representation; 2.1 Data Encoding in Neuromorphic Systems; 2.2 Spike Computing for Neuromorphic Systems.

3 Synaptic Plasticity for Information Computing3.1 Causal Approach: Synaptic Learning Versus Synaptic Adaptation; 3.2 Phenomenological Approach: Short-Term Plasticity Versus Long-Term Plasticity; 4 Synaptic Plasticity Implementation in Neuromorphic Nanodevices; 4.1 Causal Implementation of Synaptic Plasticity; 4.2 Phenomenological Implementation of Synaptic Plasticity; 5 Conclusions; References; Neuromemristive Systems: A Circuit Design Perspective; 1 Introduction: Taking a Cue from Nature; 2 Memristor Overview; 3 Voltage Versus Current-Mode Circuit Designs for NMSs.

4 Neuron Circuits: Primary Information Processing Units4.1 Input Stage; 4.2 Activation Function; 5 Synapse Circuits: Communication and Memory; 6 Plasticity Circuits: Adaptation/Learning; 7 Summary and Outlook; References; Memristor-Based Platforms: A Comparison Between Continous-Time and Discrete-Time Cellular Neural Networks; 1 Introduction; 2 Backgorund; 3 New Memristance Restoring Circuit; 4 Simulation Results; 5 Cellular Automata and DTCNNs; 6 Belief Propagation Inspired Algorithm and Cellular Automaton Equivalence for RGB Image Processing; 7 Element Detection in RGB Image; 8 Conclusions.

Multiple Binary OxRAMs as Synapses for Convolutional Neural Networks1 Multiple Binary OxRAM Devices as Artificial Synapses; 2 Convolutional Neural Network Architecture; 3 Synaptic Weight Resolution and Tolerance to Variability; 4 Conclusions; References; Nonvolatile Memory Crossbar Arrays for Non-von Neumann Computing; 1 Introduction; 2 Considerations for a Crossbar Implementation; 3 Phase-Change Memory (PCM): Results; 3.1 Experimental Results; 4 Non-filamentary RRAM Results; 4.1 Fabrication of PCMO Devices; 4.2 Simulation Results; 5 Discussion; 6 Conclusions; References.

This book covers all major aspects of cutting-edge research in the field of neuromorphic hardware engineering involving emerging nanoscale devices. Special emphasis is given to leading works in hybrid low-power CMOS-Nanodevice design. The book offers readers a bidirectional (top-down and bottom-up) perspective on designing efficient bio-inspired hardware. At the nanodevice level, it focuses on various flavors of emerging resistive memory (RRAM) technology. At the algorithm level, it addresses optimized implementations of supervised and stochastic learning paradigms such as: spike-time-dependent plasticity (STDP), long-term potentiation (LTP), long-term depression (LTD), extreme learning machines (ELM) and early adoptions of restricted Boltzmann machines (RBM) to name a few. The contributions discuss system-level power/energy/parasitic trade-offs, and complex real-world applications. The book is suited for both advanced researchers and students interested in the field.