CrossNAS: A Cross-Layer Neural Architecture Search Framework for PIM Systems
This work addresses the challenge of efficient deployment of ML workloads on PIM systems, which is incremental as it applies existing NAS strategies to a new context.
The paper tackles the problem of optimizing machine learning workloads for analog processing-in-memory (PIM) systems by proposing CrossNAS, a framework that explores a cross-layer search space and sets a new benchmark in PIM neural architecture search, outperforming previous methods in accuracy and energy efficiency.
In this paper, we propose the CrossNAS framework, an automated approach for exploring a vast, multidimensional search space that spans various design abstraction layers-circuits, architecture, and systems-to optimize the deployment of machine learning workloads on analog processing-in-memory (PIM) systems. CrossNAS leverages the single-path one-shot weight-sharing strategy combined with the evolutionary search for the first time in the context of PIM system mapping and optimization. CrossNAS sets a new benchmark for PIM neural architecture search (NAS), outperforming previous methods in both accuracy and energy efficiency while maintaining comparable or shorter search times.