Reproduce “X-OPM: Explainable Automatic Digital On-Chip Power Modeling for Enhanced Robustness” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Jingbo Jiang, Xizi Chen, Jian Peng, Wei Zhang
X-OPM estimates processor power from signal activity to guide proactive management. It derives circuit-informed features from control, configuration, and data signals, then uses a tree to propose interactions and a linear model to predict power. On held-out C906 benchmarks, it reports R² of 0.955 versus 0.908 for raw features, but saturates near 100 mW; its predictor occupies 0.61% of the core.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.