Self-supervised Learning of Image Embedding for Continuous Control

by   Carlos Florensa, et al.
berkeley college

Operating directly from raw high dimensional sensory inputs like images is still a challenge for robotic control. Recently, Reinforcement Learning methods have been proposed to solve specific tasks end-to-end, from pixels to torques. However, these approaches assume the access to a specified reward which may require specialized instrumentation of the environment. Furthermore, the obtained policy and representations tend to be task specific and may not transfer well. In this work we investigate completely self-supervised learning of a general image embedding and control primitives, based on finding the shortest time to reach any state. We also introduce a new structure for the state-action value function that builds a connection between model-free and model-based methods, and improves the performance of the learning algorithm. We experimentally demonstrate these findings in three simulated robotic tasks.


page 6

page 7


Visuomotor Control in Multi-Object Scenes Using Object-Aware Representations

Perceptual understanding of the scene and the relationship between its d...

Visual Foresight: Model-Based Deep Reinforcement Learning for Vision-Based Robotic Control

Deep reinforcement learning (RL) algorithms can learn complex robotic sk...

Simplified Temporal Consistency Reinforcement Learning

Reinforcement learning is able to solve complex sequential decision-maki...

Planning to Explore via Self-Supervised World Models

Reinforcement learning allows solving complex tasks, however, the learni...

ViewMix: Augmentation for Robust Representation in Self-Supervised Learning

Joint Embedding Architecture-based self-supervised learning methods have...

Measuring Visual Generalization in Continuous Control from Pixels

Self-supervised learning and data augmentation have significantly reduce...

Neural Algorithmic Reasoners are Implicit Planners

Implicit planning has emerged as an elegant technique for combining lear...

Please sign up or login with your details

Forgot password? Click here to reset