Presentation
Neuromorphic Swarm on RRAM Compute-in-Memory Processor for solving QUBO Problem
DescriptionCombinatorial optimization problems prevail in engineering and industry. Some are NP-hard and thus difficult to solve on edge devices. Quadratic Unconstrained Binary Optimization (QUBO) problem is an emerging model that can be used to formulate numerous combinatorial problems. In this work, we design a neuromorphic QUBO solver composed of a swarm of spiking neural networks (SNN) that conduct a meta-heuristic search for solutions. We evaluate the solver on a 40nm Resistive RAM (RRAM) Compute-in-Memory (CIM) SoC with an RRAM-based accelerator and a Cortex M3 core. Such an algorithm-hardware synergized solver is superior in speed and energy efficiency for edge devices.
Event Type
Research Manuscript
TimeWednesday, July 12th11:10am - 11:25am PDT
Location3010, 3rd Floor
Design
Emerging Models of Computation