mlsanityData to 7 Oct 2026

PushWorld: A benchmark for manipulation planning with tools and movable obstacles

arXiv 2301.10289 ↗First posted 24 Jan 2023cs.AIcited by 3

Not on the map. Why?

Abstract (arXiv)

While recent advances in artificial intelligence have achieved human-level performance in environments like Starcraft and Go, many physical reasoning tasks remain challenging for modern algorithms. To date, few algorithms have been evaluated on physical tasks that involve manipulating objects when movable obstacles are present and when tools must be used to perform the manipulation. To promote research on such tasks, we introduce PushWorld, an environment with simplistic physics that requires manipulation planning with both movable obstacles and tools. We provide a benchmark of more than 200 PushWorld puzzles in PDDL and in an OpenAI Gym environment. We evaluate state-of-the-art classical planning and reinforcement learning algorithms on this benchmark, and we find that these baseline results are below human-level performance. We then provide a new classical planning heuristic that solves the most puzzles among the baselines, and although it is 40 times faster than the best baseline planner, it remains below human-level performance.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 23 of its 70 references that are in the map. First posted on arXiv 24 Jan 2023.

Builds on

23 of 70 references.

References · 23

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441
  2. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
  3. 2019Mastering Atari, Go, Chess and Shogi by Planning with a Learned Modelcited by 1,327
  4. 2017Hindsight Experience Replaycited by 1,096
  5. 2017Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithmcited by 723
  6. 2016Unifying Count-Based Exploration and Intrinsic Motivationcited by 632
  7. 2020Curriculum Learning for Reinforcement Learning Domains: A Framework and Surveycited by 350
  8. 2016Modular Multitask Reinforcement Learning with Policy Sketchescited by 200
  9. 2016Value Iteration Networkscited by 158
  10. 2018PDDLStream: Integrating Symbolic Planners and Blackbox Samplers via Optimistic Adaptive Planningcited by 147
  11. 2020Acme: A Research Framework for Distributed Reinforcement Learningcited by 146
  12. 2017Stochastic Neural Networks for Hierarchical Reinforcement Learningcited by 124
11 more
  1. 2014The LAMA Planner: Guiding Cost-Based Anytime Planning with Landmarkscited by 86
  2. 2019Hindsight Credit Assignmentcited by 74
  3. 2018Diversity-Driven Exploration Strategy for Deep Reinforcement Learningcited by 65
  4. 2019An investigation of model-free planningcited by 64
  5. 2020Counterfactual Credit Assignment in Model-Free Reinforcement Learningcited by 60
  6. 2020Decoupling Exploration and Exploitation for Meta-Reinforcement Learning without Sacrificescited by 59
  7. 2017Schema Networks: Zero-shot Transfer with a Generative Causal Model of Intuitive Physicscited by 57
  8. 2020On the role of planning in model-based deep reinforcement learningcited by 57
  9. 2019Sub-policy Adaptation for Hierarchical Reinforcement Learningcited by 47
  10. 2022LISA: Learning Interpretable Skill Abstractions from Languagecited by 30
  11. 2021There Is No Turning Back: A Self-Supervised Approach for Reversibility-Aware Reinforcement Learningcited by 17

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 1; 2024: 1; 2025: 0; 2026: 1.

3 papers.

  1. 2023Learn to Follow: Decentralized Lifelong Multi-agent Pathfinding via Planning and Learningcited by 27
  2. 2024Graph Learning for Numeric Planningcited by 5
  3. 2026Learning to Search and Searching to Learn for Generalization in Planningcited by 0