The field of neuromorphic computing and optimization is rapidly advancing, with a focus on developing innovative hardware and algorithms that mimic the efficiency and adaptability of the human brain. Recent developments have centered around the creation of brain-inspired computing systems, such as memristor-based chaotic circuits and spin-orbit-torque magnetic content-addressable memory, which enable efficient processing of complex data streams and optimization tasks. These systems have shown significant promise in improving solution quality, speed, and energy efficiency in various applications, including the traveling salesman problem and genomic analysis. Notable papers in this area include:

• TAXI, which introduces an in-memory computing-based accelerator for the traveling salesman problem, achieving significant improvements in solution quality and energy efficiency.
• HyDra, which presents a generalized, reconfigurable on-chip training and inference architecture for hyperdimensional computing, utilizing spin-orbit-torque magnetic content-addressable memory and achieving significant energy reductions.