Corpus ID: 199668689

PHYRE: A New Benchmark for Physical Reasoning

@inproceedings{Bakhtin2019PHYREAN,
  title={PHYRE: A New Benchmark for Physical Reasoning},
  author={A. Bakhtin and L. V. D. Maaten and Justin Johnson and L. Gustafson and Ross B. Girshick},
  booktitle={NeurIPS},
  year={2019}
}
Understanding and reasoning about physics is an important ability of intelligent agents. [...] Key Method We test several modern learning algorithms on PHYRE and find that these algorithms fall short in solving the puzzles efficiently. We expect that PHYRE will encourage the development of novel sample-efficient agents that learn efficient but useful models of physics. For code and to play PHYRE for yourself, please visit this https URL.Expand
Cracking PHYRE with a World Model
  • Sheng Liu
  • 2020
Hi-Phy: A Benchmark for Hierarchical Physical Reasoning
Solving Physics Puzzles by Reasoning about Paths
Physical Reasoning Using Dynamics-Aware Models
Generalization to New Actions in Reinforcement Learning
Physion: Evaluating Physical Prediction from Vision in Humans and Machines
...
1
2
3
4
...

References

SHOWING 1-10 OF 62 REFERENCES
Physical Reasoning
  • E. Davis
  • Computer Science
  • Handbook of Knowledge Representation
  • 2008
Running the Table: An AI for Computer Billiards
The Tools Challenge: Rapid Trial-and-Error Learning in Physical Problem Solving
Probing Physics Knowledge Using Tools from Developmental Psychology
Inferring and Executing Programs for Visual Reasoning
ShapeStacks: Learning Vision-Based Physical Intuition for Generalised Object Stacking
Do New Caledonian crows solve physical problems through causal reasoning?
CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoning
...
1
2
3
4
5
...