paper-with-me

홈 › Papers

Visual Reinforcement Learning with Imagined Goals

2018-07-12 · NeurIPS 2018 12 · Ashvin Nair, Vitchyr Pong, Murtaza Dalal, Shikhar Bahl, Steven Lin, Sergey Levine

For an autonomous agent to fulfill a wide range of user-specified goals at test time, it must be able to learn broadly applicable and general-purpose skill repertoires. Furthermore, to provide the requisite level of generality, these skills must handle raw sensory input such as images. In this paper, we propose an algorithm that acquires such general-purpose skills by combining unsupervised representation learning and reinforcement learning of goal-conditioned policies. Since the particular goals that might be required at test-time are not known in advance, the agent performs a self-supervised "practice" phase where it imagines goals and attempts to achieve them. We learn a visual representation with three distinct purposes: sampling goals for self-supervised practice, providing a structured transformation of raw sensory inputs, and computing a reward signal for goal reaching. We also propose a retroactive goal relabeling scheme to further improve the sample-efficiency of our method. Our off-policy algorithm is efficient enough to learn policies that operate on raw image observations and goals for a real-world robotic system, and substantially outperforms prior techniques.

📄 PDF Abstract BibTeX arXiv:1807.04742

Code (2)

vitchyr/rlkit 공식 구현 tf
vitchyr/multiworld

Tasks

reinforcement-learningReinforcement LearningReinforcement Learning (RL)Representation Learning

Similar Papers 제목 키워드 기반

Goal-Conditioned Reinforcement Learning with Imagined Subgoals

2021-07-01 · Elliot Chane-Sane, Cordelia Schmid, Ivan Laptev

Goal-conditioned reinforcement learning endows an agent with a large variety of skills, but it often struggles to solve tasks that require more temporally extended reasoning. In this work, we propose to incorporate imagi…

reinforcement-learningReinforcement LearningReinforcement Learning (RL)

Physically-Grounded Goal Imagination: Physics-Informed Variational Autoencoder for Self-Supervised Reinforcement Learning

2025-11-10 · Lan Thi Ha Nguyen, Kien Ton Manh, Anh Do Duc, Nam Pham Hai arxiv

Self-supervised goal-conditioned reinforcement learning enables robots to autonomously acquire diverse skills without human supervision. However, a central challenge is the goal setting problem: robots must propose feasi…

Reinforcement Learning

Following Instructions by Imagining and Reaching Visual Goals

2020-01-25 · John Kanu, Eadom Dessalene, Xiaomin Lin, Cornelia Fermuller 외

While traditional methods for instruction-following typically assume prior linguistic and perceptual knowledge, many recent works in reinforcement learning (RL) have proposed learning policies end-to-end, typically by tr…

Instruction FollowingReinforcement LearningReinforcement Learning (RL)Spatial Reasoning

Electrocorticographic Dynamics Predict Visually Guided Motor Imagery of Grasp Shaping

2017-02-21

Identification of intended movement type and movement phase of hand grasp shaping are critical features for the control of volitional neuroprosthetics. We demonstrate that neural dynamics during visually-guided imagined …

Motor Imagery

Forward-Backward Reinforcement Learning

2018-03-27 · Ashley D. Edwards, Laura Downs, James C. Davidson

Goals for reinforcement learning problems are typically defined through hand-specified rewards. To design such problems, developers of learning algorithms must inherently be aware of what the task goals are, yet we often…

reinforcement-learningReinforcement LearningReinforcement Learning (RL)