Neuromorphic Circuits with Spiking Astrocytes for Increased Energy Efficiency, Fault Tolerance, and Memory Capacitance

TL;DR

Addresses robustness and energy efficiency in neuromorphic SNNs by integrating the LIFA astrocyte model. The approach couples four pillars ARR, Hopfield-inspired memory management, astrocyte based routing and fault tolerance, and LIFA hardware mapping to a fault-tolerant many-core design. Key findings include fault tolerance of 81.10% and resilience improvement of 18.90%, along with memory efficiency gains and energy savings demonstrated via CPU/GPU fault injection experiments. The work demonstrates that astrocyte-enabled neuromorphic circuits can provide robust and energy-efficient platforms for next generation neuromorphic computing.

Abstract

In the rapidly advancing field of neuromorphic computing, integrating biologically-inspired models like the Leaky Integrate-and-Fire Astrocyte (LIFA) into spiking neural networks (SNNs) enhances system robustness and performance. This paper introduces the LIFA model in SNNs, addressing energy efficiency, memory management, routing mechanisms, and fault tolerance. Our core architecture consists of neurons, synapses, and astrocyte circuits, with each astrocyte supporting multiple neurons for self-repair. This clustered model improves fault tolerance and operational efficiency, especially under adverse conditions. We developed a routing methodology to map the LIFA model onto a fault-tolerant, many-core design, optimizing network functionality and efficiency. Our model features a fault tolerance rate of 81.10\% and a resilience improvement rate of 18.90\%, significantly surpassing other implementations. The results validate our approach in memory management, highlighting its potential as a robust solution for advanced neuromorphic computing applications. The integration of astrocytes represents a significant advancement, setting the stage for more resilient and adaptable neuromorphic systems.

Figures (7)

• Figure 1: Inserting LIFA in a neural network.
• Figure 2: Self-repair mechanism of an astrocyte.
• Figure 3: Schematic representation of the LIFA model.
• Figure 4: Astrocyte Integrated Clustering.
• Figure 5: LIFA Model Energy Consumption.