End-to-end Deep Learning of Optimization Heuristics
- Submitting institution
-
The University of Lancaster
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 222875636
- Type
- E - Conference contribution
- DOI
-
10.1109/PACT.2017.24
- Title of conference / published proceedings
- The 26th International Conference on Parallel Architectures and Compilation Techniques (PACT) 2017
- First page
- 219
- Volume
- -
- Issue
- -
- ISSN
- -
- Open access status
- -
- Month of publication
- September
- Year of publication
- 2017
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
3
- Research group(s)
-
D - Distributed Systems
- Citation count
- 18
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Won the Best paper award in PACT 2017, a premier conference in parallel computing (1 out 108 submissions, top 1%). Develops an end-to-end approach for constructing compiler heuristics by learning effective representation from raw source code. Offers huge reductions in development effort and improved performance. Was the most downloaded paper of the proceedings on IEEE explore (as of March 2020). Between 2019 and 2020, it was among the top-20 most frequently accessed articles of all papers published over the last 28 years of this conference series. Helped to secure a prestigious RAEng Research Fellowship and a Royal Society collaboration grant.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
