Efficient Exploration via First-Person Behavior Cloning Assisted Rapidly-Exploring Random Trees
원문 초록
Modern day computer games have extremely large state and action spaces. To detect bugs in these games' models, human testers play the games repeatedly to explore the game and find errors in the games. Such gameplay is exhaustive and time consuming. Moreover, since robotics simulators depend on similar methods of model specification and debugging, the problem of finding errors in the model is of interest to the robotics community to ensure robot behaviors and interactions are consistent in simulators. Previous methods have used reinforcement learning arXiv:2103.13798 and search based methods (Chang, 2019, (Chang, 2021) arXiv:1811.06962 including Rapidly-exploring Random Trees (RRT) to explore a game's state-action space to find bugs. However, such search and exploration based methods are not efficient at exploring the state-action space without a pre-defined heuristic. In this work we attempt to combine a human-tester's expertise in solving games, and the RRT's exhaustiveness to search a game's state space efficiently with high coverage. This paper introduces Cloning Assisted RRT (CA-RRT) to test a game through search. We compare our methods to two existing baselines: 1) a weighted-RRT as described by arXiv:1812.03125; 2) human demonstration seeded RRT as described by Chang et. al. We find CA-RRT is applicable to more game maps and explores more game states in fewer tree expansions/iterations when compared to the existing baselines. In each test, CA-RRT reached more states on average in the same number of iterations as weighted-RRT. In our tested environments, CA-RRT reached the same number of states as weighted-RRT by more than 5000 fewer iterations on average, almost a 50% reduction and applied to more scenarios than. Moreover, as a consequence of our first person behavior cloning approach, CA-RRT worked on unseen game maps than just seeding the RRT with human demonstrated states.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
상각 샘플러를 위한 적응형 교사
Adaptive teachers for amortized samplers
상각 추론(Amortized inference)은 신경망과 같은 매개변수 모델을 학습시켜, 정확한 샘플링이 불가능한 주어진 비정규화 밀도(unnormalized density)의 분포를 근사하는 작업이다. 샘플링이 순차적 의사결정 과정으로 구현될 때, 생성 플로우 네트워크(generative flow networks)와 같은 강화학습(RL) 방법을 사용하여 샘플링 정책을 학습시킬 수 있다. 오프-정책 강화학습 학습은 다양하고 보상이 높은 후보를 발견하는 데 도움이 되지만, 기존 방법들은 효율적인 탐험에서 여전히 어려움을 겪는다. 우리는 적응형 학습 분포(\teacher)를 사용하여 주된 상각 샘플러(\student)의 학습을 안내할 것을 제안한다. 보조 행동 모델인 \teacher는 \student의 손실이 큰 영역에서 샘플링하도록 학습되며, 아직 탐험되지 않은 모드들로 일반화할 수 있어 효율적인 학습 커리큘럼을 제공함으로써 모드 커버리지(mode coverage)를 향상시킨다. 우리는 탐험 과제를 제시하도록 설계된 합성 환경, 두 가지 확산 기반 샘플링 과제, 그리고 네 가지 생화학적 발견 과제에서 이 접근법의 효과를 검증하며, 샘플 효율성과 모드 커버리지를 개선할 수 있음을 보인다. 소스 코드는 https://github.com/alstn12088/adaptive-teacher 에서 확인할 수 있다.
MermaidFlow: Redefining Agentic Workflow Generation via Safety-Constrained Evolutionary Programming
DISCOVER: Automated Curricula for Sparse-Reward Reinforcement Learning
ForesightNav: Learning Scene Imagination for Efficient Exploration
Disentangling Uncertainties by Learning Compressed Data Representation
Reward-Centered ReST-MCTS: A Robust Decision-Making Framework for Robotic Manipulation in High Uncertainty Environments
토론
이 논문과 연결된 포럼 글이 아직 없습니다.