Learning in Memristive Neural Network Architectures using Analog Backpropagation Circuits
This work addresses the need for faster training in memristive neural networks, offering a circuit-level solution that is incremental in applying existing backpropagation methods to new hardware contexts.
The paper tackled the problem of on-chip implementation of learning algorithms for neural networks by proposing analog backpropagation circuits for various memristive architectures, achieving circuit-level design and verification using TSMC 180nm CMOS and TiO2 memristor models, with validation on tasks like XOR, MNIST, and Yale face databases.
The on-chip implementation of learning algorithms would speed-up the training of neural networks in crossbar arrays. The circuit level design and implementation of backpropagation algorithm using gradient descent operation for neural network architectures is an open problem. In this paper, we proposed the analog backpropagation learning circuits for various memristive learning architectures, such as Deep Neural Network (DNN), Binary Neural Network (BNN), Multiple Neural Network (MNN), Hierarchical Temporal Memory (HTM) and Long-Short Term Memory (LSTM). The circuit design and verification is done using TSMC 180nm CMOS process models, and TiO2 based memristor models. The application level validations of the system are done using XOR problem, MNIST character and Yale face image databases