Rules, not rankings: interpretable AI for processor microarchitecture design analysis

Gamatié, A.; Yang, L.; Pimenta, P. (2026)

Article / Letter to editor

Expert-driven microarchitecture analysis provides valuable insights but fails to scale to modern design spaces, while black-box AI offers scalability at the cost of interpretability. We present PTAH, a data-driven methodology that combines codified expert knowledge with an interpretable conditional-density model. PTAH derives human-readable rules that link processor mechanism configurations to the full IPC probability distribution. Clustering these rules yields design principles that highlight key interactions, trade-offs, and behavioral regimes. A case study on SPEC CPU workloads shows that PTAH uncovers design insights while preserving full interpretability.

• All authors: Gamatié, A.; Yang, L.; Pimenta, P.
• Date: 2026-01-01
• Journal: IEEE Computer Architecture Letters
• Volume: 25
• Issue: 1
• Pages: 194 - 197
• DOI: doi:10.1109/LCA.2026.3692901
• Link: https://ieeexplore.ieee.org/document/11517228