서브메뉴
검색
Constraint Inference in Control and Reinforcement Learning
Constraint Inference in Control and Reinforcement Learning
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20250211152823
- ISBN
- 9798384449676
- DDC
- 629.8
- 서명/저자
- Constraint Inference in Control and Reinforcement Learning
- 발행사항
- [Sl] : University of California, Berkeley, 2024
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2024
- 형태사항
- 104 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 86-03, Section: B.
- 주기사항
- Advisor: Komvopoulos, Kyriakos.
- 학위논문주기
- Thesis (Ph.D.)--University of California, Berkeley, 2024.
- 초록/해제
- 요약Inferring unknown constraints is a challenging and crucial problem in many robotics applications. When only expert demonstrations are available, it becomes essential to infer the unknown domain constraints to deploy additional agents effectively. In this work, we propose approaches to infer constraints by observing experts act in an environment in different scenarios.First, we develop an approach to infer affine constraints in control tasks by observing expert demonstrations. We formulate the constraint inference problem as an inverse optimization problem, and we propose an alternating optimization scheme that infers the unknown constraints by minimizing a KKT residual objective. We demonstrate the effectiveness of our method in a number of simulations, and show that our method can infer less conservative constraints than another baseline method, while maintaining comparable safety guarantees.Second, we move to a Reinforcement Learning framework and we consider the problem of inferring constraints from demonstrations using a Bayesian perspective. We propose Bayesian Inverse Constraint Reinforcement Learning (BICRL), a novel approach that infers a posterior probability distribution over constraints from demonstrated trajectories. The main advantages of BICRL, compared to prior constraint inference algorithms, are (1) the freedom to infer constraints from partial trajectories and even from disjoint state-action pairs, (2) the ability to infer constraints from suboptimal demonstrations and in stochastic environments, and (3) the opportunity to use the posterior distribution over constraints in order to implement active learning and robust policy optimization techniques. We show that BICRL outperforms pre-existing constraint learning approaches, leading to more accurate constraint inference and consequently safer policies. We further propose Hierarchical BICRL that infers constraints locally in sub-spaces of the entire domain and then composes global constraint estimates leading to accurate and computationally efficient constraint estimation.Our third contribution is also based on a Reinforcement Learning framework. In that, we propose a novel Bayesian method that infers constraints based on preferences over demonstrations. The main advantages of our proposed approach are that it (1) infers constraints without calculating a new policy at each iteration, (2) uses a simple and more realistic ranking of groups of demonstrations, without requiring pairwise comparisons over all demonstrations, and (3) adapts to cases where there are varying levels of constraint violation. Our empirical results demonstrate that our proposed Bayesian approach infers constraints of varying severity, more accurately than state-of-the-art constraint inference methods.
- 일반주제명
- Robotics
- 일반주제명
- Computer science
- 일반주제명
- Mechanical engineering
- 키워드
- Active learning
- 기타저자
- University of California, Berkeley Mechanical Engineering
- 기본자료저록
- Dissertations Abstracts International. 86-03B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008250123s2024 us c eng d■001000017164034
■00520250211152823
■006m o d
■007cr#unu||||||||
■020 ▼a9798384449676
■035 ▼a(MiAaPQ)AAI31559862
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a629.8
■1001 ▼aPapadimitriou, Dimitris.
■24510▼aConstraint Inference in Control and Reinforcement Learning
■260 ▼a[Sl]▼bUniversity of California, Berkeley▼c2024
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2024
■300 ▼a104 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 86-03, Section: B.
■500 ▼aAdvisor: Komvopoulos, Kyriakos.
■5021 ▼aThesis (Ph.D.)--University of California, Berkeley, 2024.
■520 ▼aInferring unknown constraints is a challenging and crucial problem in many robotics applications. When only expert demonstrations are available, it becomes essential to infer the unknown domain constraints to deploy additional agents effectively. In this work, we propose approaches to infer constraints by observing experts act in an environment in different scenarios.First, we develop an approach to infer affine constraints in control tasks by observing expert demonstrations. We formulate the constraint inference problem as an inverse optimization problem, and we propose an alternating optimization scheme that infers the unknown constraints by minimizing a KKT residual objective. We demonstrate the effectiveness of our method in a number of simulations, and show that our method can infer less conservative constraints than another baseline method, while maintaining comparable safety guarantees.Second, we move to a Reinforcement Learning framework and we consider the problem of inferring constraints from demonstrations using a Bayesian perspective. We propose Bayesian Inverse Constraint Reinforcement Learning (BICRL), a novel approach that infers a posterior probability distribution over constraints from demonstrated trajectories. The main advantages of BICRL, compared to prior constraint inference algorithms, are (1) the freedom to infer constraints from partial trajectories and even from disjoint state-action pairs, (2) the ability to infer constraints from suboptimal demonstrations and in stochastic environments, and (3) the opportunity to use the posterior distribution over constraints in order to implement active learning and robust policy optimization techniques. We show that BICRL outperforms pre-existing constraint learning approaches, leading to more accurate constraint inference and consequently safer policies. We further propose Hierarchical BICRL that infers constraints locally in sub-spaces of the entire domain and then composes global constraint estimates leading to accurate and computationally efficient constraint estimation.Our third contribution is also based on a Reinforcement Learning framework. In that, we propose a novel Bayesian method that infers constraints based on preferences over demonstrations. The main advantages of our proposed approach are that it (1) infers constraints without calculating a new policy at each iteration, (2) uses a simple and more realistic ranking of groups of demonstrations, without requiring pairwise comparisons over all demonstrations, and (3) adapts to cases where there are varying levels of constraint violation. Our empirical results demonstrate that our proposed Bayesian approach infers constraints of varying severity, more accurately than state-of-the-art constraint inference methods.
■590 ▼aSchool code: 0028.
■650 4▼aRobotics
■650 4▼aComputer science
■650 4▼aMechanical engineering
■653 ▼aActive learning
■653 ▼aBayesian optimization
■653 ▼aConstraint inference problem
■653 ▼aInverse Control Problem
■653 ▼aInverse Reinforcement Learning
■653 ▼aPreference learning
■690 ▼a0771
■690 ▼a0984
■690 ▼a0548
■71020▼aUniversity of California, Berkeley▼bMechanical Engineering.
■7730 ▼tDissertations Abstracts International▼g86-03B.
■790 ▼a0028
■791 ▼aPh.D.
■792 ▼a2024
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17164034▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.


