Temporal Convolutional Networks (TCNs) are emerging lightweight Deep Learning models for Time Series analysis. We introduce an automated exploration approach and a library of optimized kernels to map TCNs on Parallel Ultra-Low Power (PULP) microcontrollers. Our approach minimizes latency and energy by exploiting a layer tiling optimizer to jointly find the tiling dimensions and select among alternative implementations of the causal and dilated 1D-convolution operations at the core of TCNs. We benchmark our approach on a commercial PULP device, achieving up to 103 x lower latency and 20.3 x lower energy than the Cube-AI toolkit executed on the STM32L4 and from 2.9 x to 26.6x lower energy compared to commercial closed-source and academic open-source approaches on the same hardware target.

TCN Mapping Optimization for Ultra-Low Power Time-Series Edge Inference

Alessio Burrello
Primo
;
Alberto Dequino
Secondo
;
Francesco Conti;Marcello Zanghieri;Luca Benini
Penultimo
;
2021

Abstract

Temporal Convolutional Networks (TCNs) are emerging lightweight Deep Learning models for Time Series analysis. We introduce an automated exploration approach and a library of optimized kernels to map TCNs on Parallel Ultra-Low Power (PULP) microcontrollers. Our approach minimizes latency and energy by exploiting a layer tiling optimizer to jointly find the tiling dimensions and select among alternative implementations of the causal and dilated 1D-convolution operations at the core of TCNs. We benchmark our approach on a commercial PULP device, achieving up to 103 x lower latency and 20.3 x lower energy than the Cube-AI toolkit executed on the STM32L4 and from 2.9 x to 26.6x lower energy compared to commercial closed-source and academic open-source approaches on the same hardware target.
2021
2021 IEEE/ACM International Symposium on Low Power Electronics and Design (ISLPED)
1
6
Alessio Burrello; Alberto Dequino; Daniele Jahier Pagliari; Francesco Conti; Marcello Zanghieri; Enrico Macii; Luca Benini; Massimo Poncino
File in questo prodotto:
Eventuali allegati, non sono esposti

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11585/904825
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 3
  • ???jsp.display-item.citation.isi??? 16
social impact