Learn to Floorplan through Acquisition of Effective Local Search Heuristics

Zhuolun He; Yuzhe Ma; Lu Zhang; Peiyu Liao; Ngai Wong; Bei Yu; Martin D. F. Wong

doi:10.1109/ICCD50377.2020.00061

Learn to Floorplan through Acquisition of Effective Local Search Heuristics

Zhuolun He, Yuzhe Ma, Lu Zhang, Peiyu Liao, Ngai Wong, Bei Yu, Martin D. F. Wong

Office of the Provost

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

24 Citations (Scopus)

Abstract

Automatic heuristic design through reinforcement learning opens a promising direction for solving computationally difficult problems. Unlike most previous works that aimed at solution construction, we explore the possibility of acquiring local search heuristics through massive search experiments. To illustrate the applicability, an agent is trained to perform a walk in the search space by selecting a candidate neighbor solution at each step. Specifically, we target the floorplanning problem, where a neighbor solution is generated through perturbing the sequence pair encoding of a floorplan. Experimental results demonstrate the efficacy of the acquired heuristics as well as the potential of automatic heuristic design.

Original language	English
Title of host publication	Proceedings of The IEEE 38th International Conference on Computer Design, ICCD 2020
Publisher	IEEE
Pages	324-331
Number of pages	8
ISBN (Electronic)	9781728197104
ISBN (Print)	9781728197111
DOIs	https://doi.org/10.1109/ICCD50377.2020.00061
Publication status	Published - 20 Oct 2020
Event	38th IEEE International Conference on Computer Design, ICCD 2020 - Hartford, United States Duration: 18 Oct 2020 → 21 Oct 2020 https://www.iccd-conf.com/2020/Home.html (Conference website) https://www.iccd-conf.com/2020/Program_2020.html (Conference programme) https://ieeexplore.ieee.org/xpl/conhome/9282809/proceeding (Conference proceedings)

Publication series

Name	Proceedings - IEEE International Conference on Computer Design (ICCD): VLSI in Computers and Processors
Volume	2020-October
ISSN (Print)	1063-6404
ISSN (Electronic)	2576-6996

Conference

Conference	38th IEEE International Conference on Computer Design, ICCD 2020
Country/Territory	United States
City	Hartford
Period	18/10/20 → 21/10/20
Internet address	https://www.iccd-conf.com/2020/Home.html (Conference website) https://www.iccd-conf.com/2020/Program_2020.html (Conference programme) https://ieeexplore.ieee.org/xpl/conhome/9282809/proceeding (Conference proceedings)

Scopus Subject Areas

Hardware and Architecture
Electrical and Electronic Engineering

User-Defined Keywords

Floorplanning
sequence pair
reinforcement learning

Access to Document

10.1109/ICCD50377.2020.00061

Cite this

He, Z., Ma, Y., Zhang, L., Liao, P., Wong, N., Yu, B., & Wong, M. D. F. (2020). Learn to Floorplan through Acquisition of Effective Local Search Heuristics. In Proceedings of The IEEE 38th International Conference on Computer Design, ICCD 2020 (pp. 324-331). (Proceedings - IEEE International Conference on Computer Design (ICCD): VLSI in Computers and Processors; Vol. 2020-October). IEEE. https://doi.org/10.1109/ICCD50377.2020.00061

@inproceedings{abe1c9c9b07a4cc8bb2eef15a4529886,

title = "Learn to Floorplan through Acquisition of Effective Local Search Heuristics",

abstract = "Automatic heuristic design through reinforcement learning opens a promising direction for solving computationally difficult problems. Unlike most previous works that aimed at solution construction, we explore the possibility of acquiring local search heuristics through massive search experiments. To illustrate the applicability, an agent is trained to perform a walk in the search space by selecting a candidate neighbor solution at each step. Specifically, we target the floorplanning problem, where a neighbor solution is generated through perturbing the sequence pair encoding of a floorplan. Experimental results demonstrate the efficacy of the acquired heuristics as well as the potential of automatic heuristic design.",

keywords = "Floorplanning, sequence pair, reinforcement learning",

author = "Zhuolun He and Yuzhe Ma and Lu Zhang and Peiyu Liao and Ngai Wong and Bei Yu and Wong, {Martin D. F.}",

note = "Funding Information: This work is partially supported by The Research Grants Council of Hong Kong SAR (No. CUHK14209420). Publisher Copyright: {\textcopyright} 2020 IEEE.; 38th IEEE International Conference on Computer Design, ICCD 2020 ; Conference date: 18-10-2020 Through 21-10-2020",

year = "2020",

month = oct,

day = "20",

doi = "10.1109/ICCD50377.2020.00061",

language = "English",

isbn = "9781728197111",

series = "Proceedings - IEEE International Conference on Computer Design (ICCD): VLSI in Computers and Processors",

publisher = "IEEE",

pages = "324--331",

booktitle = "Proceedings of The IEEE 38th International Conference on Computer Design, ICCD 2020",

address = "United States",

url = "https://www.iccd-conf.com/2020/Home.html, https://www.iccd-conf.com/2020/Program_2020.html, https://ieeexplore.ieee.org/xpl/conhome/9282809/proceeding",

}

He, Z, Ma, Y, Zhang, L, Liao, P, Wong, N, Yu, B & Wong, MDF 2020, Learn to Floorplan through Acquisition of Effective Local Search Heuristics. in Proceedings of The IEEE 38th International Conference on Computer Design, ICCD 2020. Proceedings - IEEE International Conference on Computer Design (ICCD): VLSI in Computers and Processors, vol. 2020-October, IEEE, pp. 324-331, 38th IEEE International Conference on Computer Design, ICCD 2020, Hartford, Connecticut, United States, 18/10/20. https://doi.org/10.1109/ICCD50377.2020.00061

Learn to Floorplan through Acquisition of Effective Local Search Heuristics. / He, Zhuolun; Ma, Yuzhe; Zhang, Lu et al.
Proceedings of The IEEE 38th International Conference on Computer Design, ICCD 2020. IEEE, 2020. p. 324-331 (Proceedings - IEEE International Conference on Computer Design (ICCD): VLSI in Computers and Processors; Vol. 2020-October).

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

TY - GEN

T1 - Learn to Floorplan through Acquisition of Effective Local Search Heuristics

AU - He, Zhuolun

AU - Ma, Yuzhe

AU - Zhang, Lu

AU - Liao, Peiyu

AU - Wong, Ngai

AU - Yu, Bei

AU - Wong, Martin D. F.

PY - 2020/10/20

Y1 - 2020/10/20

N2 - Automatic heuristic design through reinforcement learning opens a promising direction for solving computationally difficult problems. Unlike most previous works that aimed at solution construction, we explore the possibility of acquiring local search heuristics through massive search experiments. To illustrate the applicability, an agent is trained to perform a walk in the search space by selecting a candidate neighbor solution at each step. Specifically, we target the floorplanning problem, where a neighbor solution is generated through perturbing the sequence pair encoding of a floorplan. Experimental results demonstrate the efficacy of the acquired heuristics as well as the potential of automatic heuristic design.

AB - Automatic heuristic design through reinforcement learning opens a promising direction for solving computationally difficult problems. Unlike most previous works that aimed at solution construction, we explore the possibility of acquiring local search heuristics through massive search experiments. To illustrate the applicability, an agent is trained to perform a walk in the search space by selecting a candidate neighbor solution at each step. Specifically, we target the floorplanning problem, where a neighbor solution is generated through perturbing the sequence pair encoding of a floorplan. Experimental results demonstrate the efficacy of the acquired heuristics as well as the potential of automatic heuristic design.

KW - Floorplanning

KW - sequence pair

KW - reinforcement learning

UR - http://www.scopus.com/inward/record.url?scp=85098843749&partnerID=8YFLogxK

U2 - 10.1109/ICCD50377.2020.00061

DO - 10.1109/ICCD50377.2020.00061

M3 - Conference proceeding

AN - SCOPUS:85098843749

SN - 9781728197111

T3 - Proceedings - IEEE International Conference on Computer Design (ICCD): VLSI in Computers and Processors

SP - 324

EP - 331

BT - Proceedings of The IEEE 38th International Conference on Computer Design, ICCD 2020

PB - IEEE

T2 - 38th IEEE International Conference on Computer Design, ICCD 2020

Y2 - 18 October 2020 through 21 October 2020

ER -

He Z, Ma Y, Zhang L, Liao P, Wong N, Yu B et al. Learn to Floorplan through Acquisition of Effective Local Search Heuristics. In Proceedings of The IEEE 38th International Conference on Computer Design, ICCD 2020. IEEE. 2020. p. 324-331. (Proceedings - IEEE International Conference on Computer Design (ICCD): VLSI in Computers and Processors). doi: 10.1109/ICCD50377.2020.00061

Learn to Floorplan through Acquisition of Effective Local Search Heuristics

Abstract

Publication series

Conference

Scopus Subject Areas

User-Defined Keywords

Access to Document

Other files and links

Fingerprint

Cite this