Home > Research > Publications & Outputs > BOSS: Bayesian Optimization over String Spaces

Research

Associated organisational units

Electronic data

  • BOSS

    2.4 MB, PDF document

    Available under license: CC BY: Creative Commons Attribution 4.0 International License

Links

View graph of relations

BOSS: Bayesian Optimization over String Spaces

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review

Published

Standard

BOSS: Bayesian Optimization over String Spaces. / Moss, Henry; Beck, Daniel; Gonzalez, Javier et al.
Advances in Neural Information Processing Systems. Vol. 2020-December 2020. ed. MIT Press, 2020. p. 15476-15486 (Advances in Neural Information Processing Systems).

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review

Harvard

Moss, H, Beck, D, Gonzalez, J, Leslie, D & Rayson, P 2020, BOSS: Bayesian Optimization over String Spaces. in Advances in Neural Information Processing Systems. 2020 edn, vol. 2020-December, Advances in Neural Information Processing Systems, MIT Press, pp. 15476-15486, 34th Conference on Neural Information Processing Systems, Vancouver, Canada, 6/12/20. <https://nips.cc/Conferences/2020>

APA

Moss, H., Beck, D., Gonzalez, J., Leslie, D., & Rayson, P. (2020). BOSS: Bayesian Optimization over String Spaces. In Advances in Neural Information Processing Systems (2020 ed., Vol. 2020-December, pp. 15476-15486). (Advances in Neural Information Processing Systems). MIT Press. https://nips.cc/Conferences/2020

Vancouver

Moss H, Beck D, Gonzalez J, Leslie D, Rayson P. BOSS: Bayesian Optimization over String Spaces. In Advances in Neural Information Processing Systems. 2020 ed. Vol. 2020-December. MIT Press. 2020. p. 15476-15486. (Advances in Neural Information Processing Systems).

Author

Moss, Henry ; Beck, Daniel ; Gonzalez, Javier et al. / BOSS: Bayesian Optimization over String Spaces. Advances in Neural Information Processing Systems. Vol. 2020-December 2020. ed. MIT Press, 2020. pp. 15476-15486 (Advances in Neural Information Processing Systems).

Bibtex

@inproceedings{47f0f248ada04df6a541c170fb727e1a,
title = "BOSS: Bayesian Optimization over String Spaces",
abstract = "This article develops a Bayesian optimization (BO) method which acts directly over raw strings, proposing the first uses of string kernels and genetic algorithms within BO loops. Recent applications of BO over strings have been hindered by the need to map inputs into a smooth and unconstrained latent space. Learning this projection is computationally and data-intensive. Our approach instead builds a powerful Gaussian process surrogate model based on string kernels, naturally supporting variable length inputs, and performs efficient acquisition function maximization for spaces with syntactical constraints. Experiments demonstrate considerably improved optimization over existing approaches across a broad range of constraints, including the popular setting where syntax is governed by a context-free grammar.",
author = "Henry Moss and Daniel Beck and Javier Gonzalez and David Leslie and Paul Rayson",
year = "2020",
month = dec,
day = "1",
language = "English",
isbn = "9780262042062",
volume = "2020-December",
series = "Advances in Neural Information Processing Systems",
publisher = "MIT Press",
pages = "15476--15486",
booktitle = "Advances in Neural Information Processing Systems",
edition = "2020",
note = "34th Conference on Neural Information Processing Systems, NeurIPS 2020 ; Conference date: 06-12-2020 Through 12-12-2020",
url = "https://nips.cc/Conferences/2020",

}

RIS

TY - GEN

T1 - BOSS: Bayesian Optimization over String Spaces

AU - Moss, Henry

AU - Beck, Daniel

AU - Gonzalez, Javier

AU - Leslie, David

AU - Rayson, Paul

PY - 2020/12/1

Y1 - 2020/12/1

N2 - This article develops a Bayesian optimization (BO) method which acts directly over raw strings, proposing the first uses of string kernels and genetic algorithms within BO loops. Recent applications of BO over strings have been hindered by the need to map inputs into a smooth and unconstrained latent space. Learning this projection is computationally and data-intensive. Our approach instead builds a powerful Gaussian process surrogate model based on string kernels, naturally supporting variable length inputs, and performs efficient acquisition function maximization for spaces with syntactical constraints. Experiments demonstrate considerably improved optimization over existing approaches across a broad range of constraints, including the popular setting where syntax is governed by a context-free grammar.

AB - This article develops a Bayesian optimization (BO) method which acts directly over raw strings, proposing the first uses of string kernels and genetic algorithms within BO loops. Recent applications of BO over strings have been hindered by the need to map inputs into a smooth and unconstrained latent space. Learning this projection is computationally and data-intensive. Our approach instead builds a powerful Gaussian process surrogate model based on string kernels, naturally supporting variable length inputs, and performs efficient acquisition function maximization for spaces with syntactical constraints. Experiments demonstrate considerably improved optimization over existing approaches across a broad range of constraints, including the popular setting where syntax is governed by a context-free grammar.

UR - http://www.scopus.com/inward/record.url?scp=85099807878&partnerID=8YFLogxK

M3 - Conference contribution/Paper

SN - 9780262042062

VL - 2020-December

T3 - Advances in Neural Information Processing Systems

SP - 15476

EP - 15486

BT - Advances in Neural Information Processing Systems

PB - MIT Press

T2 - 34th Conference on Neural Information Processing Systems

Y2 - 6 December 2020 through 12 December 2020

ER -