AutoTinyML for microcontrollers: Dealing with black-box deployability. (30th November 2022)

• Record Type: Journal Article
• Authors: Perego, Riccardo; Candelieri, Antonio; Archetti, Francesco; Pau, Danilo
• Is Part Of: Expert systems with applications. Volume 207(2022)
• Publication Date: 2022-11-30
• Subjects: Tiny Machine Learning -- Deep Neural Networks -- Automated Machine Learning -- Neural Architecture Search -- Hyperparameter optimization -- Bayesian Optimization; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2022.117876
• Languages: English
• ISSNs: 0957-4174

Abstract

Abstract: While many companies are currently leveraging on Cloud, data centres and specialized hardware (e.g., GPUs and TPUs) to train very accurate Machine Learning models, the need to deploy and run these models on tiny devices is emerging as the most relevant challenge, with a massive untapped market. Although Automated Machine Learning and Neural Architecture Search frameworks are successfully used to find accurate models by trying a small number of alternatives, they are typically performed on large computational platforms and they cannot directly deal with deployability, leading to an accurate model which could result undeployable on a tiny device. To bridge the gap between these two worlds, we present an approach extending these frameworks to include the constraints related to the limited hardware resources of the tiny device which the trained model has to run on. Experimental results on two benchmark classification tasks and two microcontrollers prove that our AutoTinyML framework can efficiently identify models which are both accurate and deployable, in case accepting a reasonable reduction in accuracy compared to a significant reduction in hardware usages, without applying any quantization techniques of the model. Highlights: Constrained Bayesian Optimization includes deployment constraints for an HPO problem Deep Neural Networks deployable on MCUs is a massive untapped market for TinyML Finding out most accurate DNN taking into account deployment constraints.