Peter Stone's Selected Publications

Classified by TopicClassified by Publication TypeSorted by DateSorted by First Author Last NameClassified by Funding Source


Efficient Real-Time Inference in Temporal Convolution Networks

Efficient Real-Time Inference in Temporal Convolution Networks.
Piyush Khandelwal, James MacGlashan, Peter Wurman, and Peter Stone.
In Proceedings of the 2021 International Conference on Robotics and Automation (ICRA 2021), May 2021.

Download

[PDF]346.4kB  

Abstract

It has been recently demonstrated that Temporal Convolution Networks (TCNs) provide state-of-the-art results in many problem domains where the input data is a time-series. TCNs typically incorporate information from a long history of inputs (the receptive field) into a single output using many convolution layers. Real-time inference using a trained TCN can be challenging on devices with limited compute and memory, especially if the receptive field is large. This paper introduces the RT-TCN algorithm that reuses the output of prior convo- lution operations to minimize the computational requirements and persistent memory footprint of a TCN during real-time inference. We also show that when a TCN is trained using time slices of the input time-series, it can be executed in real- time continually using RT-TCN. In addition, we provide TCN architecture guidelines that ensure that real-time inference can be performed within memory and computational constraints.

BibTeX Entry

@inProceedings {ICRA21-Piyush,
	author = {Piyush Khandelwal and James MacGlashan and Peter Wurman and Peter Stone},
	title = {Efficient Real-Time Inference in Temporal Convolution Networks},
	booktitle = {Proceedings of the 2021 International Conference on Robotics and Automation (ICRA 2021)},
	location = {Xi’an China},
	month = {May},
	year = {2021},
	abstract = {
           It has been recently demonstrated that Temporal Convolution
           Networks (TCNs) provide state-of-the-art results in many
           problem domains where the input data is a time-series.
           TCNs typically incorporate information from a long history
           of inputs (the receptive field) into a single output using
           many convolution layers. Real-time inference using a
           trained TCN can be challenging on devices with limited
           compute and memory, especially if the receptive field is
           large. This paper introduces the RT-TCN algorithm that
           reuses the output of prior convo- lution operations to
           minimize the computational requirements and persistent
           memory footprint of a TCN during real-time inference. We
           also show that when a TCN is trained using time slices of
           the input time-series, it can be executed in real- time
           continually using RT-TCN. In addition, we provide TCN
           architecture guidelines that ensure that real-time
           inference can be performed within memory and computational
           constraints.
	},
}

Generated by bib2html.pl (written by Patrick Riley ) on Wed Nov 17, 2021 22:19:00