Work

Chameleon: Adaptive Code Optimization for Expedited Deep Neural Network Compilation

Year: 2020

Type: conference-paper

Abstract: provides 4.45×speed up in optimization time over AutoTVM, while also improving inference time of the modern deep networks by 5.6%.

Source: PubMed

DOI: 10.48550/arxiv.2001.08743

Authors: Ahn, Byung Hoon, Pilligundla, Prannoy, Yazdanbakhsh, Amir, Esmaeilzadeh, Hadi

Institution: University of California San Diego

Language: English

Topic: Parallel Computing and Optimization Techniques

Subfield: Hardware and Architecture

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funder: National Institute of Biomedical Imaging and Bioengineering

Award: A framework for developing translatable intelligent neural interface systems for precision neuromodulation therapies

OpenAlex ID: W2995723163

FWCI: 6.173

Cites: 44

Cited by: 42

Related to: 20