Improving Robustness of ReRAM-based Spiking Neural Network Accelerator with Stochastic Spike-timing-dependent-plasticity
This work addresses robustness issues in neuromorphic computing accelerators for applications like pattern recognition, though it appears incremental as it builds on existing STDP methods.
The paper tackled the problem of input noise and device variation in ReRAM-based spiking neural network accelerators using STDP, by introducing a novel stochastic STDP algorithm that improved accuracy in pattern recognition tasks with noisy inputs over deterministic STDP and enhanced resilience to device variation.
Spike-timing-dependent-plasticity (STDP) is an unsupervised learning algorithm for spiking neural network (SNN), which promises to achieve deeper understanding of human brain and more powerful artificial intelligence. While conventional computing system fails to simulate SNN efficiently, process-in-memory (PIM) based on devices such as ReRAM can be used in designing fast and efficient STDP based SNN accelerators, as it operates in high resemblance with biological neural network. However, the real-life implementation of such design still suffers from impact of input noise and device variation. In this work, we present a novel stochastic STDP algorithm that uses spiking frequency information to dynamically adjust synaptic behavior. The algorithm is tested in pattern recognition task with noisy input and shows accuracy improvement over deterministic STDP. In addition, we show that the new algorithm can be used for designing a robust ReRAM based SNN accelerator that has strong resilience to device variation.