TinyNS: Platform-aware Neurosymbolic Auto Tiny Machine Learning
ACM Transactions on Embedded Computing Systems · 31.05.2024 · 23 (3) · 1-48
DOI: 10.1145/3603171 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM Transactions on Embedded Computing Systems, online erschienen am 11.05.2024. Verfasst von Swapnil Sayan Saha und 6 weiteren Autor:innen. Beteiligte Einrichtung: University of California - Los Angeles, Los Angeles, CA, USA. Die Arbeit wurde bisher 26-mal zitiert (Crossref) und nennt 180 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIBIB NIH HHS, Air Force Office of Scientific Research (AFOSR) under Cooperative Agreement, IoBT REIGN Collaborative Research Alliance funded by the Army Research Laboratory (ARL) under Cooperative Agreement.
Abstract (Originalsprache)
Machine learning at the extreme edge has enabled a plethora of intelligent, time-critical, and remote applications. However, deploying interpretable artificial intelligence systems that can perform high-level symbolic reasoning and satisfy the underlying system rules and physics within the tight platform resource constraints is challenging. In this article, we introduce TinyNS , the first platform-aware neurosymbolic architecture search framework for joint optimization of symbolic and neural operators. TinyNS provides recipes and parsers to automatically write microcontroller code for five types of neurosymbolic models, combining the context awareness and integrity of symbolic techniques with the robustness and performance of machine learning models. TinyNS uses a fast, gradient-free, black-box Bayesian optimizer over discontinuous, conditional, numeric, and categorical search spaces to find the best synergy of symbolic code and neural networks within the hardware resource budget. To guarantee deployability, TinyNS talks to the target hardware during the optimization process. We showcase the utility of TinyNS by deploying microcontroller-class neurosymbolic models through several case studies. In all use cases, TinyNS outperforms purely neural or purely symbolic approaches while guaranteeing execution on real hardware.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- University of California - Los Angeles, Los Angeles, CA, USA.
- Abacus.AI, Seattle, WA, USA.
- BrightNight, Austin, TX, USA.
Förderung
- NIBIB NIH HHS
- Air Force Office of Scientific Research (AFOSR) under Cooperative Agreement
- IoBT REIGN Collaborative Research Alliance funded by the Army Research Laboratory (ARL) under Cooperative Agreement
- NIH mHealth Center for Discovery, Optimization and Translation of Temporally Precise Interventions
- National Science Foundation
- CONIX Research Center, one of six centers in JUMP, a Semiconductor Research Corporation (SRC) program
Zitieren
APA
Saha, S. S., Sandha, S. S., Aggarwal, M., Wang, B., Han, L., DE Gortari Briseno, J., & Srivastava, M. (2024). TinyNS: Platform-aware Neurosymbolic Auto Tiny Machine Learning. ACM Transactions on Embedded Computing Systems, 23(3), 1-48. https://doi.org/10.1145/3603171
Vancouver
Saha SS, Sandha SS, Aggarwal M, Wang B, Han L, DE Gortari Briseno J, et al. TinyNS: Platform-aware Neurosymbolic Auto Tiny Machine Learning. ACM Transactions on Embedded Computing Systems. 2024;23(3):1-48. doi:10.1145/3603171
BibTeX
@article{Saha2024,
author = {Saha, Swapnil Sayan and Sandha, Sandeep Singh and Aggarwal, Mohit and Wang, Brian and Han, Liying and DE Gortari Briseno, Julian and Srivastava, Mani},
title = {TinyNS: Platform-aware Neurosymbolic Auto Tiny Machine Learning},
journal = {ACM Transactions on Embedded Computing Systems},
year = {2024},
volume = {23},
number = {3},
pages = {1--48},
doi = {10.1145/3603171}
}
EUVIMED