State Representation

State representation in artificial intelligence focuses on creating effective internal models of an environment's state, enabling agents to make informed decisions. Current research emphasizes learning these representations automatically, often using contrastive learning, deep reinforcement learning, and large language models (LLMs) to extract meaningful features from high-dimensional data, including images and text, and to improve generalization and sample efficiency. These advancements are crucial for improving the robustness and adaptability of AI agents in complex, dynamic environments, with applications ranging from robotics and recommender systems to autonomous driving and natural language processing.

Papers

September 20, 2022

Locally Constrained Representations in Reinforcement Learning
Somjit Nath, Rushiv Arora, Samira Ebrahimi Kahou
Reinforcement Learning Policy Learning Robust Representation State Representation Sufficient Representation

September 12, 2022

Unified State Representation Learning under Data Augmentation
Taylor Hearn, Sravan Jayanthi, Sehoon Ha
Reinforcement Learning Domain Adaptation Data Augmentation State Representation

September 6, 2022

Project proposal: A modular reinforcement learning based automated theorem prover
Boris Shminke
Reinforcement Learning Theorem Proving State Representation Proposal Balance Refinement Modular Reinforcement Learning

August 26, 2022

Visual processing in context of reinforcement learning
Hlynur Davíð Hlynsson
Reinforcement Learning Deep Reinforcement Learning Context Information State Representation Visual Processing State Reward

August 8, 2022

Generating Coherent Narratives by Learning Dynamic and Discrete Entity States with a Contrastive Framework
Jian Guan, Zhenyu Yang, Rongsheng Zhang, Zhipeng Hu, Minlie Huang
Story Generation State Representation Contrastive Method Spoken Narrative Entity Knowledge Story Point Level State

July 22, 2022

Graph-Structured Policy Learning for Multi-Goal Manipulation Tasks
David Klee, Ondrej Biza, Robert Platt
Q Learning Robotic Manipulation Manipulation Task State Representation Multi Goal Reinforcement Learning

July 15, 2022

Bootstrap State Representation using Style Transfer for Better Generalization in Deep Reinforcement Learning
Md Masudur Rahman, Yexiang Xue
Deep Reinforcement Learning Strong Generalization Style Transfer Policy Learning Better Generalization State Representation Human Thinking

June 17, 2022

CtrlFormer: Learning Transferable State Representation for Visual Control via Transformer
Yao Mu, Shoufa Chen, Mingyu Ding, Jianyu Chen, Runjian Chen, Ping Luo
Transformer Based State Representation Visual Control Contrastive Reinforcement Learning

June 3, 2022

Reinforcement Learning with Neural Radiance Fields
Danny Driess, Ingmar Schubert, Pete Florence, Yunzhu Li, Marc Toussaint
Reinforcement Learning Neural Radiance Field Latent Space State Representation

May 4, 2022

State Representation Learning for Goal-Conditioned Reinforcement Learning
Lorenzo Steccanella, Anders Jonsson
State Representation Goal Conditioned Reinforcement Learning Goal Conditioned Policy Reward Free Reinforcement Learning Multi Goal

April 2, 2022

Moment-based Adversarial Training for Embodied Language Comprehension
Shintaro Ishikawa, Komei Sugiura
Adversarial Training Robot Person Vision Language Task State Representation

March 7, 2022

The Unsurprising Effectiveness of Pre-Trained Vision Models for Control
Simone Parisi, Aravind Rajeswaran, Senthil Purushwalkam, Abhinav Gupta
Pre Trained Large Scale External Control Policy Learning Pre Trained Representation State Representation Surprising Effectiveness

March 4, 2022

Symbolic State Estimation with Predicates for Contact-Rich Manipulation Tasks
Toki Migimatsu, Wenzhao Lian, Jeannette Bohg, Stefan Schaal
State Estimation Behavior Tree State Representation Contact Rich Manipulation Task Lex Leader Symmetry Breaking Predicate Peg in Hole Finite State

March 1, 2022

On the Generalization of Representations in Reinforcement Learning
Charline Le Lan, Stephen Tu, Adam Oberman, Rishabh Agarwal, Marc G. Bellemare
Reinforcement Learning Strong Generalization Meaningful Representation State Representation Successor Representation Generalization Behavior

January 28, 2022

Mask-based Latent Reconstruction for Reinforcement Learning
Tao Yu, Zhizheng Zhang, Cuiling Lan, Yan Lu, Zhibo Chen
Reinforcement Learning Deep Reinforcement Learning State Representation Masked Reconstruction Latent Mask

January 18, 2022

Accelerating Representation Learning with View-Consistent Dynamics in Data-Efficient Reinforcement Learning
Tao Huang, Jiachen Wang, Xiao Chen
Deep Reinforcement Learning Representation Learning Data Efficient State Representation

December 31, 2021

SimSR: Simple Distance-based State Representation for Deep Reinforcement Learning
Hongyu Zang, Xin Li, Mingzhong Wang
Deep Reinforcement Learning State Representation Multi Agent MuJoCo Representation Collapse

December 15, 2021

Feature-Attending Recurrent Modules for Generalization in Reinforcement Learning
Wilka Carvalho, Andrew Lampinen, Kyriacos Nikiforou, Felix Hill, Murray Shanahan
Reinforcement Learning Strong Generalization State Representation Recurrent Module Object Centric Task Object Centric Generative Model

November 24, 2021

Learning State Representations via Retracing in Reinforcement Learning
Changmin Yu, Dong Li, Jianye Hao, Jun Wang, Neil Burgess
Reinforcement Learning State Representation Model Learning Reinforcement Learning Task Retraction Based Method