Compute-in-Memory Implementation of State Space Models for Event Sequence Processing
This work addresses the problem of energy-efficient, real-time sequence processing for event-driven applications, representing an incremental improvement through algorithm-hardware co-design.
The authors tackled the challenge of implementing state space models (SSMs) for real-time event-driven processing by developing a compute-in-memory (CIM) hardware approach, achieving high accuracy and energy efficiency for event-based vision and audio tasks.
State space models (SSMs) have recently emerged as a powerful framework for long sequence processing, outperforming traditional methods on diverse benchmarks. Fundamentally, SSMs can generalize both recurrent and convolutional networks and have been shown to even capture key functions of biological systems. Here we report an approach to implement SSMs in energy-efficient compute-in-memory (CIM) hardware to achieve real-time, event-driven processing. Our work re-parameterizes the model to function with real-valued coefficients and shared decay constants, reducing the complexity of model mapping onto practical hardware systems. By leveraging device dynamics and diagonalized state transition parameters, the state evolution can be natively implemented in crossbar-based CIM systems combined with memristors exhibiting short-term memory effects. Through this algorithm and hardware co-design, we show the proposed system offers both high accuracy and high energy efficiency while supporting fully asynchronous processing for event-based vision and audio tasks.