English

LRMP: Layer Replication with Mixed Precision for Spatial In-memory DNN Accelerators

Hardware Architecture 2023-12-07 v1

Abstract

In-memory computing (IMC) with non-volatile memories (NVMs) has emerged as a promising approach to address the rapidly growing computational demands of Deep Neural Networks (DNNs). Mapping DNN layers spatially onto NVM-based IMC accelerators achieves high degrees of parallelism. However, two challenges that arise in this approach are the highly non-uniform distribution of layer processing times and high area requirements. We propose LRMP, a method to jointly apply layer replication and mixed precision quantization to improve the performance of DNNs when mapped to area-constrained NVM-based IMC accelerators. LRMP uses a combination of reinforcement learning and integer linear programming to search the replication-quantization design space using a model that is closely informed by the target hardware architecture. Across five DNN benchmarks, LRMP achieves 2.8-9×\times latency and 11.8-19×\times throughput improvement at iso-accuracy.

Keywords

Cite

@article{arxiv.2312.03146,
  title  = {LRMP: Layer Replication with Mixed Precision for Spatial In-memory DNN Accelerators},
  author = {Abinand Nallathambi and Christin David Bose and Wilfried Haensch and Anand Raghunathan},
  journal= {arXiv preprint arXiv:2312.03146},
  year   = {2023}
}
R2 v1 2026-06-28T13:42:17.128Z