OpenACMv2: An Accuracy-Constrained Co-Optimization Framework for Approximate DCiM
This work addresses the problem of efficient hardware design for neural network acceleration in approximate DCiM systems, though it appears incremental as it builds on prior tools like OpenYield.
The paper tackles the challenge of optimizing power-performance-area (PPA) in approximate Digital Compute-in-Memory (DCiM) while maintaining accuracy constraints, by introducing OpenACMv2, a framework that uses a two-level optimization approach to achieve significant PPA improvements under controlled accuracy budgets.
Digital Compute-in-Memory (DCiM) accelerates neural networks by reducing data movement. Approximate DCiM can further improve power-performance-area (PPA), but demands accuracy-constrained co-optimization across coupled architecture and transistor-level choices. Building on OpenYield, we introduce Accuracy-Constrained Co-Optimization (ACCO) and present OpenACMv2, an open framework that operationalizes ACCO via two-level optimization: (1) accuracy-constrained architecture search of compressor combinations and SRAM macro parameters, driven by a fast GNN-based surrogate for PPA and error; and (2) variation- and PVT-aware transistor sizing for standard cells and SRAM bitcells using Monte Carlo. By decoupling ACCO into architecture-level exploration and circuit-level sizing, OpenACMv2 integrates classic single- and multi-objective optimizers to deliver strong PPA-accuracy tradeoffs and robust convergence. The workflow is compatible with FreePDK45 and OpenROAD, supporting reproducible evaluation and easy adoption. Experiments demonstrate significant PPA improvements under controlled accuracy budgets, enabling rapid "what-if" exploration for approximate DCiM. The framework is available on https://github.com/ShenShan123/OpenACM.