mlsanityData to 7 Oct 2026

Rotation, Translation, and Cropping for Zero-Shot Generalization

arXiv 2001.09908 ↗First posted 27 Jan 2020cs.LG cs.CV stat.MLcited by 23

Not on the map. Why?

Abstract (arXiv)

Deep Reinforcement Learning (DRL) has shown impressive performance on domains with visual inputs, in particular various games. However, the agent is usually trained on a fixed environment, e.g. a fixed number of levels. A growing mass of evidence suggests that these trained models fail to generalize to even slight variations of the environments they were trained on. This paper advances the hypothesis that the lack of generalization is partly due to the input representation, and explores how rotation, cropping and translation could increase generality. We show that a cropped, translated and rotated observation can get better generalization on unseen levels of two-dimensional arcade games from the GVGAI framework. The generality of the agents is evaluated on both human-designed and procedurally generated levels.

Key facts. Cited by 23 papers in the map (data through 7 Oct 2026). First posted on arXiv 27 Jan 2020.

Builds on

No reference list yet.

Built on by

Cited by year

Citing papers per year: 2021: 5; 2022: 8; 2023: 4; 2024: 4; 2025: 2.

23 papers.

Sort:

By area:

  1. 2021Collaborating with Humans without Human Datacited by 139
  2. 2021Contrastive Behavioral Similarity Embeddings for Generalization in Reinforcement Learningcited by 123
  3. 2021Decoupling Value and Policy for Generalization in Reinforcement Learningcited by 81
  4. 2023On the Importance of Exploration for Generalization in Reinforcement Learningcited by 30
  5. 2023The Generalization Gap in Offline Reinforcement Learningcited by 25
  6. 2023Explore to Generalize in Zero-Shot RLcited by 20
  7. 2022Learning Task-relevant Representations for Generalization via Characteristic Functions of Reward Sequence Distributionscited by 15
  8. 2022Path of Destruction: Learning an Iterative Level Generator Using a Small Datasetcited by 14
  9. 2022Warmth and competence in human-agent cooperationcited by 11
  10. 2024The Overcooked Generalisation Challenge: Evaluating Cooperation with Novel Partners in Unknown Environments Using Unsupervised Environment Designcited by 11
  11. 2022Mutation Models: Learning to Generate Levels by Imitating Evolutioncited by 9
  12. 2021Tuning Mixed Input Hyperparameters on the Fly for Efficient Population Based AutoRLcited by 8
11 more
  1. 2024PCGRL+: Scaling, Control and Generalization in Reinforcement Learning Level Generatorscited by 7
  2. 2022Transfer Dynamics in Emergent Evolutionary Curriculacited by 6
  3. 2024Improving Generalization on the ProcGen Benchmark with Simple Architectural Changes and Scalecited by 6
  4. 2022Learning Generalizable Representations for Reinforcement Learning via Adaptive Meta-learner of Behavioral Similaritiescited by 4
  5. 2023Equivariant Data Augmentation for Generalization in Offline Reinforcement Learningcited by 3
  6. 2025Provable Zero-Shot Generalization in Offline Reinforcement Learningcited by 3
  7. 2022Predicting Personas Using Mechanic Frequencies and Game State Tracescited by 2
  8. 2022ARLO: A Framework for Automated Reinforcement Learningcited by 2
  9. 2025Adaptive Data Exploitation in Deep Reinforcement Learningcited by 1
  10. 2021Active Reinforcement Learning over MDPscited by 0
  11. 2024Do Agents Dream of Electric Sheep?: Improving Generalization in Reinforcement Learning through Generative Learningcited by 0