Learning Temporally-Consistent Representations for Data-Efficient Reinforcement Learning

by   Trevor McInroe, et al.

Deep reinforcement learning (RL) agents that exist in high-dimensional state spaces, such as those composed of images, have interconnected learning burdens. Agents must learn an action-selection policy that completes their given task, which requires them to learn a representation of the state space that discerns between useful and useless information. The reward function is the only supervised feedback that RL agents receive, which causes a representation learning bottleneck that can manifest in poor sample efficiency. We present k-Step Latent (KSL), a new representation learning method that enforces temporal consistency of representations via a self-supervised auxiliary task wherein agents learn to recurrently predict action-conditioned representations of the state space. The state encoder learned by KSL produces low-dimensional representations that make optimization of the RL task more sample efficient. Altogether, KSL produces state-of-the-art results in both data efficiency and asymptotic performance in the popular PlaNet benchmark suite. Our analyses show that KSL produces encoders that generalize better to new tasks unseen during training, and its representations are more strongly tied to reward, are more invariant to perturbations in the state space, and move more smoothly through the temporal axis of the RL problem than other methods such as DrQ, RAD, CURL, and SAC-AE.


page 5

page 15

page 18

page 20

page 21


Mask-based Latent Reconstruction for Reinforcement Learning

For deep reinforcement learning (RL) from pixels, learning effective sta...

Analyzing the Hidden Activations of Deep Policy Networks: Why Representation Matters

We analyze the hidden activations of neural network policies of deep rei...

Return-Based Contrastive Representation Learning for Reinforcement Learning

Recently, various auxiliary tasks have been proposed to accelerate repre...

Towards Governing Agent's Efficacy: Action-Conditional β-VAE for Deep Transparent Reinforcement Learning

We tackle the blackbox issue of deep neural networks in the settings of ...

Representation Learning for Continuous Action Spaces is Beneficial for Efficient Policy Learning

Deep reinforcement learning (DRL) breaks through the bottlenecks of trad...

Light-weight probing of unsupervised representations for Reinforcement Learning

Unsupervised visual representation learning offers the opportunity to le...

Reinforcement Learning from Multiple Sensors via Joint Representations

In many scenarios, observations from more than one sensor modality are a...

Please sign up or login with your details

Forgot password? Click here to reset