Reproduce “Energy Measurement and Optimization of TinyML in Resource-Constrained Embedded Systems: A Systematic Mapping Study” 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
Clariele Almeida, Renan Beserra, Rafael Moura, Danilo Araújo, Bruno Nogueira, Sérgio Chevtchenko, Ermeson Andrade
The authors map energy measurement, reported influences, and optimization approaches across 25 energy-aware TinyML studies published from 2020 through 2025, focusing mainly on inference and deployment. They find energy depends on interactions among models, hardware, communication, precision, timing, and inputs, while study methods vary. In one people-counting study, the fastest setup used 93.5 J per 30-second cycle versus 2.45 J on an Arduino; this is not a general device ranking.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.