Learning State Representations From Random Deep Action-conditional Predictions

Abstract

Our main contribution in this work is an empirical finding that random General Value Functions (GVFs), i.e., deep action-conditional predictions -- random both in what feature of observations they predict as well as in the sequence of actions the predictions are conditioned upon -- form good auxiliary tasks for reinforcement learning (RL) problems. In particular, we show that random deep action-conditional predictions when used as auxiliary tasks yield state representations that produce control performance competitive with state-of-the-art hand-crafted auxiliary tasks like value prediction, pixel control, and CURL in both Atari and DeepMind Lab tasks. In another set of experiments we stop the gradients from the RL part of the network to the state representation learning part of the network and show, perhaps surprisingly, that the auxiliary tasks alone are sufficient to learn state representations good enough to outperform an end-to-end trained actor-critic baseline. We opensourced our

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyzheng2021learning

Related papers

• Value-consistent Representation Learning For Data-efficient Reinforcement Learning (2022)0.00
• Return-based Contrastive Representation Learning For Reinforcement Learning (2021)12.17
• Towards Governing Agent's Efficacy: Action-conditional \(\beta\)-vae For Deep Transparent Reinforcement Learning (2018)0.00
• Learning To Identify Critical States For Reinforcement Learning From Videos (2023)8.76
• Bootstrapped Representations In Reinforcement Learning (2023)0.00
• Visual Processing In Context Of Reinforcement Learning (2022)0.00
• Learning Temporally-consistent Representations For Data-efficient Reinforcement Learning (2021)0.00
• Contrastive Learning As Goal-conditioned Reinforcement Learning (2022)0.00