Corpus ID: 49313355

Transfer Learning for Related Reinforcement Learning Tasks via Image-to-Image Translation

  title={Transfer Learning for Related Reinforcement Learning Tasks via Image-to-Image Translation},
  author={Shani Gamrian and Y. Goldberg},
  • Shani Gamrian, Y. Goldberg
  • Published 2019
  • Computer Science
  • ArXiv
  • Despite the remarkable success of Deep RL in learning control policies from raw pixels, the resulting models do not generalize. We demonstrate that a trained agent fails completely when facing small visual changes, and that fine-tuning---the common transfer learning paradigm---fails to adapt to these changes, to the extent that it is faster to re-train the model from scratch. We show that by separating the visual transfer task from the control policy we achieve substantially better sample… CONTINUE READING
    30 Citations
    Measuring Visual Generalization in Continuous Control from Pixels
    • PDF
    Generalization in Reinforcement Learning by Soft Data Augmentation
    • PDF
    Mutual Information Based Knowledge Transfer Under State-Action Dimension Mismatch
    • PDF
    Off-Dynamics Reinforcement Learning: Training for Transfer with Domain Classifiers
    • 2
    • PDF
    Visual Analogies between Atari Games for Studying Transfer Learning in RL
    • 2
    • PDF


    DARLA: Improving Zero-Shot Transfer in Reinforcement Learning
    • 198
    • PDF
    Zero-Shot Task Generalization with Multi-Task Deep Reinforcement Learning
    • 141
    • PDF
    Schema Networks: Zero-shot Transfer with a Generative Causal Model of Intuitive Physics
    • 144
    • PDF
    Reinforcement Learning from Imperfect Demonstrations
    • 100
    • PDF
    DualGAN: Unsupervised Dual Learning for Image-to-Image Translation
    • 973
    • PDF
    Learning from Demonstrations for Real World Reinforcement Learning
    • 108
    • PDF
    One-Shot Reinforcement Learning for Robot Navigation with Interactive Replay
    • 36
    • PDF
    Deep Learning of Representations for Unsupervised and Transfer Learning
    • Yoshua Bengio
    • Computer Science
    • ICML Unsupervised and Transfer Learning
    • 2012
    • 839
    • PDF
    Generative Adversarial Imitation Learning
    • 941
    • PDF