Sample-efficient Reinforcement Learning Representation Learning with Curiosity Contrastive Forward Dynamics Model

Abstract

Developing an agent in reinforcement learning (RL) that is capable of performing complex control tasks directly from high-dimensional observation such as raw pixels is yet a challenge as efforts are made towards improving sample efficiency and generalization. This paper considers a learning framework for Curiosity Contrastive Forward Dynamics Model (CCFDM) in achieving a more sample-efficient RL based directly on raw pixels. CCFDM incorporates a forward dynamics model (FDM) and performs contrastive learning to train its deep convolutional neural network-based image encoder (IE) to extract conducive spatial and temporal information for achieving a more sample efficiency for RL. In addition, during training, CCFDM provides intrinsic rewards, produced based on FDM prediction error, encourages the curiosity of the RL agent to improve exploration. The diverge and less-repetitive observations provide by both our exploration strategy and data augmentation available in contrastive learning improve not only the sample efficiency but also the generalization. Performance of existing model-free RL methods such as Soft Actor-Critic built on top of CCFDM outperforms prior state-of-the-art pixel-based RL methods on the DeepMind Control Suite benchmark.

References in corpus (33)

• Auto-Encoding Variational Bayes
• Continuous control with deep reinforcement learning
• A Simple Framework for Contrastive Learning of Visual Representations
• Representation Learning with Contrastive Predictive Coding
• Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
• Improved Baselines with Momentum Contrastive Learning
• End-to-End Training of Deep Visuomotor Policies
• Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model
• The Arcade Learning Environment: An Evaluation Platform for General Agents
• Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm
• QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulation
• Solving Rubik's Cube with a Robot Hand
• DeepMind Control Suite
• Recurrent World Models Facilitate Policy Evolution
• Model-Based Reinforcement Learning for Atari
• Large-Scale Study of Curiosity-Driven Learning
• Count-Based Exploration with Neural Density Models
• Learning Latent Dynamics for Planning from Pixels
• #Exploration: A Study of Count-Based Exploration for Deep Reinforcement Learning
• Visual Reinforcement Learning with Imagined Goals
• Reinforcement Learning with Unsupervised Auxiliary Tasks
• Exploration by Random Network Distillation
• Reinforcement Learning with Augmented Data
• Imagination-Augmented Agents for Deep Reinforcement Learning
• dm_control: Software and Tasks for Continuous Control
• Leveraging Procedural Generation to Benchmark Reinforcement Learning
• Model-Based Value Estimation for Efficient Model-Free Reinforcement Learning
• Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixels
• Deep Attention Recurrent Q-Network
• Stochastic Latent Actor-Critic: Deep Reinforcement Learning with a Latent Variable Model
• Learning to Perform Physics Experiments via Deep Reinforcement Learning
• Curiosity-Driven Experience Prioritization via Density Estimation
• DORA The Explorer: Directed Outreaching Reinforcement Action-Selection

Cited by in corpus (4)

• CCLF: A Contrastive-Curiosity-Driven Learning Framework for Sample-Efficient Reinforcement Learning
• DimCL: Dimensional Contrastive Learning For Improving Self-Supervised Learning
• GDCA: GAN-based single image super resolution with Dual discriminators and Channel Attention
• Metric-Driven Reinforcement Learning for Object Hallucination Reduction in LVLM-Based Image Captioning