Implementation of multilayer perceptron network with highly uniform passive memristive crossbar circuits

Abstract

The progress in the field of neural computation hinges on the use of hardware more efficient than the conventional microprocessors. Recent works have shown that mixed-signal integrated memristive circuits, especially their passive (0T1R) variety, may increase the neuromorphic network performance dramatically, leaving far behind their digital counterparts. The major obstacle, however, is immature memristor technology so that only limited functionality has been reported. Here we demonstrate operation of one-hidden layer perceptron classifier entirely in the mixed-signal integrated hardware, comprised of two passive 20 × 20 metal-oxide memristive crossbar arrays, board-integrated with discrete conventional components. The demonstrated network, whose hardware complexity is almost 10× higher as compared to previously reported functional classifier circuits based on passive memristive crossbars, achieves classification fidelity within 3% of that obtained in simulations, when using ex-situ training. The successful demonstration was facilitated by improvements in fabrication technology of memristors, specifically by lowering variations in their I–V characteristics.

Recommended Methods to Study Resistive Switching Devices

“…For such purpose, some reports built planar [54,55] and 3D [56] cross-bar arrays (see Figure 1c). [57] Note that the term cross-bar refers to a collection of interconnected cross-point devices; therefore, using the term cross-bar to refer to a single and isolated crosspoint structure is misleading. The test setup for advanced cross-bar circuit structures typically requires die packaging and dedicated printed circuit board or custom probe card in combination with switch matrix tool.…”

Bridging Biological and Artificial Neural Networks with Emerging Neuromorphic Devices: Fundamentals, Progress, and Challenges

“…[3,[27][28][29][30] In the hardware implementation of FC networks, e.g., multi-layer perceptrons (MLPs), the weight matrices could be directly mapped to the conductance matrices of memristive crossbar arrays. [3,[27][28][29][30] In the hardware implementation of FC networks, e.g., multi-layer perceptrons (MLPs), the weight matrices could be directly mapped to the conductance matrices of memristive crossbar arrays.…”

Мемристоры на основе поли-n-ксилилена с внедренными наночастицами серебра

“…Ключевые слова: мемристор, органическая электроника, поли-п-ксилилен, нанокомпозит. DOI: 10.21883/PJTF.2020.02.48948.18064 В последнее время существенно возрос интерес к структурам, демонстрирующим эффекты обратимого резистивного переключения (РП), благодаря перспективности их использования для создания элементов многоуровневой памяти и массивов мемристоров, эмулирующих синапсы в нейроморфных (биоподобных) вычислительных системах (НВС) для решения так называемых антропоморфных задач [1][2][3]: распознавания образов и естественного языка, принятия решений, обобщения, прогнозирования и др. Такого рода системы обладают малым энергопотреблением и могут существенно превосходить по эффективности обработки информации современные вычислительные системы, базирующиеся на архитектуре фон Неймана, при выполнении перечисленных выше когнитивных функций [4].…”