Adaptive Neurally-Inspired Computing: Models, Algorithms, and Silicon-Based Architectures

EIA-0130705 -华盛顿大学-Guang R. Gao-自适应神经启发计算：模型、算法和基于硅的架构在过去几十年中，硅技术的快速发展使数字设备能够在数值计算中实现超高速。然而，当遇到编程时无法预见的情况时，大多数这些设备都容易发生灾难性故障。赋予这些设备适应和从经验中学习的能力正在迅速成为信息技术中一个至关重要的问题。我们提出了一种新的解决方法：建立基于神经生物学计算和学习的信息技术系统，我们打算通过开发大脑中选定的感觉和运动区域的神经元和神经元网络的可塑性和信息处理的计算模型来实现这一目标;在基于软件的模拟中测试这些模型及其相应的算法，并使用突触晶体管和现场可编程学习阵列（FPLA）在硅中设计这些算法的实时实现。我们希望我们的研究能更好地理解神经网络中的计算，并导致新一代的自适应神经形态设备，可用于各种信息技术应用，从信号处理和模式识别到无处不在的计算和机器人控制。

EIA-0130705 -University of Washington-Guang R. Gao-Adaptive Neurally-Inspired Computing: Models, Algorithms, and Silicon-Based ArchitecturesRapid advances in silicon technology over the past few decades have allowed digitaldevices to achieve ultra-high speeds in numerical computation. However, a majority ofthese devices are prone to catastrophic failure when confronted with circumstancesunforeseen at programming time. Endowing such devices with the ability to adapt andlearn from experience is rapidly becoming a problem of fundamental importance ininformation technology. We propose a new approach to solving this problem: buildinginformation technology systems based on neurobiological computation and learning.We intend to achieve this goal by developing computational models of plasticity and information processing in neurons and networks of neurons in selected sensory and motor areas of the brain; testing these models and their corresponding algorithms in software-based simulations, and designing real-time implementations of these algorithms in silicon using synapse transistors and field-programmable learning arrays (FPLAs).We expect our research to provide a better understanding of computation within neuronalnetworks, and to lead to a new generation of adaptive neuromorphic devices that could beused for a variety of information technology applications, ranging from signal processingand pattern recognition to ubiquitous computing and robotic control.