Human-machine collaborative optimization via apprenticeship scheduling

Coordinating agents to complete a set of tasks with intercoupled temporal and resource constraints is computationally challenging, yet human domain experts can solve these difficult scheduling problems using paradigms learned through years of apprenticeship. A process for manually codifying this dom...

Full description

Bibliographic Details
Main Authors: Gombolay, Matthew C., Jensen, Reed E., Stigile, Jessica L., Golen, Toni, Shah, Neel, Son, Sung-Hyun, Shah, Julie A
Other Authors: Lincoln Laboratory
Format: Article
Language:English
Published: AI Access Foundation 2020
Online Access:https://hdl.handle.net/1721.1/125878
_version_ 1826188142748631040
author Gombolay, Matthew C.
Jensen, Reed E.
Stigile, Jessica L.
Golen, Toni
Shah, Neel
Son, Sung-Hyun
Shah, Julie A
author2 Lincoln Laboratory
author_facet Lincoln Laboratory
Gombolay, Matthew C.
Jensen, Reed E.
Stigile, Jessica L.
Golen, Toni
Shah, Neel
Son, Sung-Hyun
Shah, Julie A
author_sort Gombolay, Matthew C.
collection MIT
description Coordinating agents to complete a set of tasks with intercoupled temporal and resource constraints is computationally challenging, yet human domain experts can solve these difficult scheduling problems using paradigms learned through years of apprenticeship. A process for manually codifying this domain knowledge within a computational framework is necessary to scale beyond the "single-expert, single-trainee" apprenticeship model. However, human domain experts often have difficulty describing their decision-making processes. We propose a new approach for capturing this decision-making process through counterfactual reasoning in pairwise comparisons. Our approach is model-free and does not require iterating through the state space. We demonstrate that this approach accurately learns multifaceted heuristics on a synthetic and real world data sets. We also demonstrate that policies learned from human scheduling demonstration via apprenticeship learning can substantially improve the efficiency of schedule optimization. We employ this human-machine collaborative optimization technique on a variant of the weapon-to-target assignment problem. We demonstrate that this technique generates optimal solutions up to 9.5 times faster than a state-of-the-art optimization algorithm.
first_indexed 2024-09-23T07:55:13Z
format Article
id mit-1721.1/125878
institution Massachusetts Institute of Technology
language English
last_indexed 2024-09-23T07:55:13Z
publishDate 2020
publisher AI Access Foundation
record_format dspace
spelling mit-1721.1/1258782022-09-30T01:00:14Z Human-machine collaborative optimization via apprenticeship scheduling Gombolay, Matthew C. Jensen, Reed E. Stigile, Jessica L. Golen, Toni Shah, Neel Son, Sung-Hyun Shah, Julie A Lincoln Laboratory Coordinating agents to complete a set of tasks with intercoupled temporal and resource constraints is computationally challenging, yet human domain experts can solve these difficult scheduling problems using paradigms learned through years of apprenticeship. A process for manually codifying this domain knowledge within a computational framework is necessary to scale beyond the "single-expert, single-trainee" apprenticeship model. However, human domain experts often have difficulty describing their decision-making processes. We propose a new approach for capturing this decision-making process through counterfactual reasoning in pairwise comparisons. Our approach is model-free and does not require iterating through the state space. We demonstrate that this approach accurately learns multifaceted heuristics on a synthetic and real world data sets. We also demonstrate that policies learned from human scheduling demonstration via apprenticeship learning can substantially improve the efficiency of schedule optimization. We employ this human-machine collaborative optimization technique on a variant of the weapon-to-target assignment problem. We demonstrate that this technique generates optimal solutions up to 9.5 times faster than a state-of-the-art optimization algorithm. 2020-06-19T14:12:47Z 2020-06-19T14:12:47Z 2018-09 2019-11-01T12:12:29Z Article http://purl.org/eprint/type/JournalArticle 1943-5037 https://hdl.handle.net/1721.1/125878 Gombolay, Matthew, et al., "Human-machine collaborative optimization via apprenticeship scheduling." Journal of Artificial Intelligence Research 63 (Sept. 2018): p. 1-49 doi 10.1613/JAIR.1.11233 ©2018 Author(s) en 10.1613/JAIR.1.11233 Journal of Artificial Intelligence Research Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf AI Access Foundation arXiv
spellingShingle Gombolay, Matthew C.
Jensen, Reed E.
Stigile, Jessica L.
Golen, Toni
Shah, Neel
Son, Sung-Hyun
Shah, Julie A
Human-machine collaborative optimization via apprenticeship scheduling
title Human-machine collaborative optimization via apprenticeship scheduling
title_full Human-machine collaborative optimization via apprenticeship scheduling
title_fullStr Human-machine collaborative optimization via apprenticeship scheduling
title_full_unstemmed Human-machine collaborative optimization via apprenticeship scheduling
title_short Human-machine collaborative optimization via apprenticeship scheduling
title_sort human machine collaborative optimization via apprenticeship scheduling
url https://hdl.handle.net/1721.1/125878
work_keys_str_mv AT gombolaymatthewc humanmachinecollaborativeoptimizationviaapprenticeshipscheduling
AT jensenreede humanmachinecollaborativeoptimizationviaapprenticeshipscheduling
AT stigilejessical humanmachinecollaborativeoptimizationviaapprenticeshipscheduling
AT golentoni humanmachinecollaborativeoptimizationviaapprenticeshipscheduling
AT shahneel humanmachinecollaborativeoptimizationviaapprenticeshipscheduling
AT sonsunghyun humanmachinecollaborativeoptimizationviaapprenticeshipscheduling
AT shahjuliea humanmachinecollaborativeoptimizationviaapprenticeshipscheduling