Homeostatic Fault Tolerance in Spiking Neural Networks: A Dynamic Hardware Perspective

Abstract

Fault tolerance is a remarkable feature of biological systems and their self-repair capability influence modern electronic systems. In this paper, we propose a novel plastic neural network model, which establishes homeostasis in a spiking neural network. Combined with this plasticity and the inspiration from inhibitory interneurons, we develop a fault-resilient robotic controller implemented on an FPGA establishing obstacle avoidance task. We demonstrate the proposed methodology on a spiking neural network implemented on Xilinx Artix-7 FPGA. The system is able to maintain stable firing (tolerance ±10%) with a loss of up to 75% of the original synaptic inputs to a neuron. Our repair mechanism has minimal hardware overhead with a tuning circuit (repair unit) which consumes only three slices/neuron for implementing a threshold voltagebased homeostatic fault-tolerant unit. The overall architecture has a minimal impact on power consumption and, therefore, supports scalable implementations. This paper opens a novel way of implementing the behavior of natural fault tolerant system in hardware establishing homeostatic self-repair behavior.

Spiking Neural Networks Hardware Implementations and Challenges

“…For instance, the principles of evolution and natural selection gave rise to evolutionary hardware for implementation of reconfigurable fault tolerant systems [10,58,151]. Closer to the brain, we have seen researchers exploit neuroplasticity mechanisms to realize efficient fault tolerant and reconfigurable systems [83,109,110,166]. Neuroplasticity refers to the ability of the brain to adapt and reconfigure during lifetime operation.…”

Improved Integrate-and-Fire Neuron Models for Inference Acceleration of Spiking Neural Networks

“…The input firing rate has been proven to have an important impact on the spiking activity of SNN [8,27,28]. In this part, we 425 study the detailed impact of the input firing rate, typically, we present the spiking activities within the initial 100ms of SCNN as shown in Fig.9.…”

Fault-Tolerant Learning in Spiking Astrocyte-Neural Networks on FPGAs

“…Once this is established successfully, we apply the concept to a real-world task. There are some works which demonstrate the application of FPGAbased neural networks in solving real world tasks [19]- [21]. Compared to these works, the proposed system has enhanced fault-tolerance and learning capabilities.…”