Filters








6 Hits in 2.4 sec

MicroNets: Neural Network Architectures for Deploying TinyML Applications on Commodity Microcontrollers [article]

Colby Banbury, Chuteng Zhou, Igor Fedorov, Ramon Matas Navarro, Urmish Thakker, Dibakar Gope, Vijay Janapa Reddi, Matthew Mattina, Paul N. Whatmough
2021 arXiv   pre-print
., the mapping from a given neural network architecture to its inference latency/energy on an MCU.  ...  Executing machine learning workloads locally on resource constrained microcontrollers (MCUs) promises to drastically expand the application space of IoT.  ...  ACKNOWLEDGEMENTS This work was sponsored in part by the ADA (Applications Driving Architectures) Center.  ... 
arXiv:2010.11267v6 fatcat:cte3gwj2wnh3nlg3rnonvbpazu

Machine Learning for Microcontroller-Class Hardware – A Review [article]

Swapnil Sayan Saha, Sandeep Singh Sandha, Mani Srivastava
2022 arXiv   pre-print
We characterize a closed-loop widely applicable workflow of machine learning model development for microcontroller class devices and show that several classes of applications adopt a specific instance  ...  Conventional machine learning deployment has high memory and compute footprint hindering their direct deployment on ultra resource-constrained microcontrollers.  ...  Secondly, most NAS frameworks, TinyML software suites, intermittent computing tools, and online learning frameworks lack support for deploying some of these models on commodity microcontrollers.  ... 
arXiv:2205.14550v3 fatcat:y272riitirhwfgfiotlwv5i7nu

AnalogNets: ML-HW Co-Design of Noise-robust TinyML Models and Always-On Analog Compute-in-Memory Accelerator [article]

Chuteng Zhou, Fernando Garcia Redondo, Julian Büchel, Irem Boybat, Xavier Timoneda Comas, S. R. Nandakumar, Shidhartha Das, Abu Sebastian, Manuel Le Gallo, Paul N. Whatmough
2021 arXiv   pre-print
Always-on TinyML perception tasks in IoT applications require very high energy efficiency.  ...  This work describes AnalogNets: TinyML models for the popular always-on applications of keyword spotting (KWS) and visual wake words (VWW).  ...  This accuracy gap highlights the lack of research on model architectures and training methodologies for analog CiM. TinyML on MCUs MCUs are currently the commodity platform for TinyML.  ... 
arXiv:2111.06503v1 fatcat:tcrphrxtxredzmhwfxyjfnn5ku

Differentiable Network Pruning for Microcontrollers [article]

Edgar Liberis, Nicholas D. Lane
2021 arXiv   pre-print
Embedded and personal IoT devices are powered by microcontroller units (MCUs), whose extreme resource scarcity is a major obstacle for applications relying on on-device deep learning inference.  ...  Orders of magnitude less storage, memory and computational capacity, compared to what is typically required to execute neural networks, impose strict structural constraints on the network architecture  ...  Micronets: Neural network architectures for deploy- Lai, L., Suda, N., and Chandra, V. CMSIS-NN: Efficient ing tinyml applications on commodity microcontrollers.  ... 
arXiv:2110.08350v2 fatcat:7jmlafbgzjexbbtdyslmlx6ig4

An Overview of Machine Learning within Embedded and Mobile Devices–Optimizations and Applications

Taiwo Samuel Ajani, Agbotiname Lucky Imoize, Aderemi A. Atayero
2021 Sensors  
networks (DNNs).  ...  Embedded systems technology is undergoing a phase of transformation owing to the novel advancements in computer architecture and the breakthroughs in machine learning applications.  ...  [143] presented a survey where neural network architectures (MicroNets) target commodity microcontroller units.  ... 
doi:10.3390/s21134412 pmid:34203119 fatcat:dxmshp4frnf4pcookdy3wjl4fi

TinyOdom

Swapnil Sayan Saha, Sandeep Singh Sandha, Luis Antonio Garcia, Mani Srivastava
2022 Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies  
In this paper, we introduce TinyOdom, a framework for training and deploying neural inertial models on URC hardware.  ...  TinyOdom exploits hardware and quantization-aware Bayesian neural architecture search (NAS) and a temporal convolutional network (TCN) backbone to train lightweight models targetted towards URC devices  ...  We also thank Jason Wu from the Networked and Embedded Systems Laboratory at the University of California -Los Angeles for aiding us in the data collection phase during the real-world setup.  ... 
doi:10.1145/3534594 fatcat:n2nbtuf73nbl3jg7qj2cmd36mq