Cooperative Bayesian Optimization for Imperfect Agents

Ali Khoshvishkaie; Petrus Mikkola; Pierre Alexandre Murena; Samuel Kaski

doi:10.1007/978-3-031-43412-9_28

Cooperative Bayesian Optimization for Imperfect Agents

Ali Khoshvishkaie, Petrus Mikkola, Pierre Alexandre Murena, Samuel Kaski

Machine Learning and Robotics

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

Abstract

We introduce a cooperative Bayesian optimization problem for optimizing black-box functions of two variables where two agents choose together at which points to query the function but have only control over one variable each. This setting is inspired by human-AI teamwork, where an AI-assistant helps its human user solve a problem, in this simplest case, collaborative optimization. We formulate the solution as sequential decision-making, where the agent we control models the user as a computationally rational agent with prior knowledge about the function. We show that strategic planning of the queries enables better identification of the global maximum of the function as long as the user avoids excessive exploration. This planning is made possible by using Bayes Adaptive Monte Carlo planning and by endowing the agent with a user model that accounts for conservative belief updates and exploratory sampling of the points to query.

Original language	English
Title of host publication	Machine Learning and Knowledge Discovery in Databases: Research Track
Subtitle of host publication	European Conference, ECML PKDD 2023, Proceedings
Editors	Danai Koutra, Claudia Plant, Manuel Gomez Rodrigues, Elena Baralis, Francesco Bronchi
Place of Publication	Cham
Publisher	Springer Cham
Pages	475-490
Number of pages	16
ISBN (Electronic)	9783031434129
ISBN (Print)	9783031434112
DOIs	https://doi.org/10.1007/978-3-031-43412-9_28
Publication status	Published - 17 Sept 2023
Event	European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases - , Italy Duration: 18 Sept 2023 → 22 Sept 2023

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	14169
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases
Country/Territory	Italy
Period	18/09/23 → 22/09/23

Keywords

Bayesian optimization

Research Beacons, Institutes and Platforms

Digital Futures
Institute for Data Science and AI
Christabel Pankhurst Institute

Access to Document

10.1007/978-3-031-43412-9_28

MCAIF: Centre for AI Fundamentals
Kaski, S. (PI), Alvarez, M. (Researcher), Pan, W. (Researcher), Mu, T. (Researcher), Rivasplata, O. (PI), Sun, M. (PI), Mukherjee, A. (PI), Caprio, M. (PI), Sonee, A. (Researcher), Leroy, A. (Researcher), Wang, J. (Researcher), Lee, J. (Researcher), Parakkal Unni, M. (Researcher), Sloman, S. (Researcher), Menary, S. (Researcher), Quilter, T. (Researcher), Hosseinzadeh, A. (PGR student), Mousa, A. (PGR student), Glover, E. (PGR student), Das, A. (PGR student), DURSUN, F. (PGR student), Zhu, H. (PGR student), Abdi, H. (PGR student), Dandago, K. (PGR student), Piriyajitakonkij, M. (PGR student), Rachman, R. (PGR student), Shi, X. (PGR student), Keany, T. (PGR student), Liu, X. (PGR student), Jiang, Y. (PGR student), Wan, Z. (PGR student), Harrison, M. (Support team), Machado, M. (Support team), Hartford, J. (PI), Kangin, D. (Researcher), Harikumar, H. (PI), Dubey, M. (PI), Parakkal Unni, M. (PI), Dash, S. P. (PGR student), Mi, X. (PGR student) & Barlas, Y. (PGR student)
1/10/21 → 30/09/26
Project: Research
Turing AI Fellowship: Human-AI Research Teams - Steering AI in Experimental Design and Decision-Making
Kaski, S. (PI), Bristow, R. (CoI), Cai, P. (CoI), Jay, C. (CoI) & Peek, N. (CoI)
1/10/21 → 30/09/26
Project: Research

Cite this

Khoshvishkaie, A., Mikkola, P., Murena, P. A., & Kaski, S. (2023). Cooperative Bayesian Optimization for Imperfect Agents. In D. Koutra, C. Plant, M. G. Rodrigues, E. Baralis, & F. Bronchi (Eds.), Machine Learning and Knowledge Discovery in Databases: Research Track: European Conference, ECML PKDD 2023, Proceedings (pp. 475-490). (Lecture Notes in Computer Science; Vol. 14169). Springer Cham. https://doi.org/10.1007/978-3-031-43412-9_28

Khoshvishkaie, Ali ; Mikkola, Petrus ; Murena, Pierre Alexandre et al. / Cooperative Bayesian Optimization for Imperfect Agents. Machine Learning and Knowledge Discovery in Databases: Research Track: European Conference, ECML PKDD 2023, Proceedings. editor / Danai Koutra ; Claudia Plant ; Manuel Gomez Rodrigues ; Elena Baralis ; Francesco Bronchi. Cham : Springer Cham, 2023. pp. 475-490 (Lecture Notes in Computer Science).

@inproceedings{64fc91da1ae74326adf1973be40e0419,

title = "Cooperative Bayesian Optimization for Imperfect Agents",

abstract = "We introduce a cooperative Bayesian optimization problem for optimizing black-box functions of two variables where two agents choose together at which points to query the function but have only control over one variable each. This setting is inspired by human-AI teamwork, where an AI-assistant helps its human user solve a problem, in this simplest case, collaborative optimization. We formulate the solution as sequential decision-making, where the agent we control models the user as a computationally rational agent with prior knowledge about the function. We show that strategic planning of the queries enables better identification of the global maximum of the function as long as the user avoids excessive exploration. This planning is made possible by using Bayes Adaptive Monte Carlo planning and by endowing the agent with a user model that accounts for conservative belief updates and exploratory sampling of the points to query.",

keywords = "Bayesian optimization",

author = "Ali Khoshvishkaie and Petrus Mikkola and Murena, {Pierre Alexandre} and Samuel Kaski",

year = "2023",

month = sep,

day = "17",

doi = "10.1007/978-3-031-43412-9_28",

language = "English",

isbn = "9783031434112",

series = "Lecture Notes in Computer Science",

publisher = "Springer Cham",

pages = "475--490",

editor = "Koutra, {Danai } and Plant, {Claudia } and Rodrigues, {Manuel Gomez} and Baralis, {Elena } and Francesco Bronchi",

booktitle = "Machine Learning and Knowledge Discovery in Databases: Research Track",

address = "Switzerland",

note = "European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases ; Conference date: 18-09-2023 Through 22-09-2023",

}

Khoshvishkaie, A, Mikkola, P, Murena, PA & Kaski, S 2023, Cooperative Bayesian Optimization for Imperfect Agents. in D Koutra, C Plant, MG Rodrigues, E Baralis & F Bronchi (eds), Machine Learning and Knowledge Discovery in Databases: Research Track: European Conference, ECML PKDD 2023, Proceedings. Lecture Notes in Computer Science, vol. 14169, Springer Cham, Cham, pp. 475-490, European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Italy, 18/09/23. https://doi.org/10.1007/978-3-031-43412-9_28

Cooperative Bayesian Optimization for Imperfect Agents. / Khoshvishkaie, Ali; Mikkola, Petrus; Murena, Pierre Alexandre et al.
Machine Learning and Knowledge Discovery in Databases: Research Track: European Conference, ECML PKDD 2023, Proceedings. ed. / Danai Koutra; Claudia Plant; Manuel Gomez Rodrigues; Elena Baralis; Francesco Bronchi. Cham: Springer Cham, 2023. p. 475-490 (Lecture Notes in Computer Science; Vol. 14169).

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Cooperative Bayesian Optimization for Imperfect Agents

AU - Khoshvishkaie, Ali

AU - Mikkola, Petrus

AU - Murena, Pierre Alexandre

AU - Kaski, Samuel

PY - 2023/9/17

Y1 - 2023/9/17

N2 - We introduce a cooperative Bayesian optimization problem for optimizing black-box functions of two variables where two agents choose together at which points to query the function but have only control over one variable each. This setting is inspired by human-AI teamwork, where an AI-assistant helps its human user solve a problem, in this simplest case, collaborative optimization. We formulate the solution as sequential decision-making, where the agent we control models the user as a computationally rational agent with prior knowledge about the function. We show that strategic planning of the queries enables better identification of the global maximum of the function as long as the user avoids excessive exploration. This planning is made possible by using Bayes Adaptive Monte Carlo planning and by endowing the agent with a user model that accounts for conservative belief updates and exploratory sampling of the points to query.

AB - We introduce a cooperative Bayesian optimization problem for optimizing black-box functions of two variables where two agents choose together at which points to query the function but have only control over one variable each. This setting is inspired by human-AI teamwork, where an AI-assistant helps its human user solve a problem, in this simplest case, collaborative optimization. We formulate the solution as sequential decision-making, where the agent we control models the user as a computationally rational agent with prior knowledge about the function. We show that strategic planning of the queries enables better identification of the global maximum of the function as long as the user avoids excessive exploration. This planning is made possible by using Bayes Adaptive Monte Carlo planning and by endowing the agent with a user model that accounts for conservative belief updates and exploratory sampling of the points to query.

KW - Bayesian optimization

U2 - 10.1007/978-3-031-43412-9_28

DO - 10.1007/978-3-031-43412-9_28

M3 - Conference contribution

SN - 9783031434112

T3 - Lecture Notes in Computer Science

SP - 475

EP - 490

BT - Machine Learning and Knowledge Discovery in Databases: Research Track

A2 - Koutra, Danai

A2 - Plant, Claudia

A2 - Rodrigues, Manuel Gomez

A2 - Baralis, Elena

A2 - Bronchi, Francesco

PB - Springer Cham

CY - Cham

T2 - European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases

Y2 - 18 September 2023 through 22 September 2023

ER -

Khoshvishkaie A, Mikkola P, Murena PA, Kaski S. Cooperative Bayesian Optimization for Imperfect Agents. In Koutra D, Plant C, Rodrigues MG, Baralis E, Bronchi F, editors, Machine Learning and Knowledge Discovery in Databases: Research Track: European Conference, ECML PKDD 2023, Proceedings. Cham: Springer Cham. 2023. p. 475-490. (Lecture Notes in Computer Science). doi: 10.1007/978-3-031-43412-9_28

Cooperative Bayesian Optimization for Imperfect Agents

Abstract

Publication series

Conference

Keywords

Research Beacons, Institutes and Platforms

Access to Document

Fingerprint

Projects

MCAIF: Centre for AI Fundamentals

Turing AI Fellowship: Human-AI Research Teams - Steering AI in Experimental Design and Decision-Making

Cite this