Selected article for: "novel training scheme and training scheme"

Author: Wang, Yufei; Narasimhan, Gautham Narayan; Lin, Xingyu; Okorn, Brian; Held, David
Title: ROLL: Visual Self-Supervised Reinforcement Learning with Object Reasoning
  • Cord-id: z3emsqxy
  • Document date: 2020_11_13
  • ID: z3emsqxy
    Snippet: Current image-based reinforcement learning (RL) algorithms typically operate on the whole image without performing object-level reasoning. This leads to inefficient goal sampling and ineffective reward functions. In this paper, we improve upon previous visual self-supervised RL by incorporating object-level reasoning and occlusion reasoning. Specifically, we use unknown object segmentation to ignore distractors in the scene for better reward computation and goal generation; we further enable occ
    Document: Current image-based reinforcement learning (RL) algorithms typically operate on the whole image without performing object-level reasoning. This leads to inefficient goal sampling and ineffective reward functions. In this paper, we improve upon previous visual self-supervised RL by incorporating object-level reasoning and occlusion reasoning. Specifically, we use unknown object segmentation to ignore distractors in the scene for better reward computation and goal generation; we further enable occlusion reasoning by employing a novel auxiliary loss and training scheme. We demonstrate that our proposed algorithm, ROLL (Reinforcement learning with Object Level Learning), learns dramatically faster and achieves better final performance compared with previous methods in several simulated visual control tasks. Project video and code are available at https://sites.google.com/andrew.cmu.edu/roll.

    Search related documents:
    Co phrase search for related documents
    • activation function and log likelihood: 1
    • activation function and log likelihood loss: 1
    • activation function and loss function: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12