Partial Observability

Partial observability, where agents lack complete information about their environment's state, is a central challenge in many areas of artificial intelligence, particularly reinforcement learning. Current research focuses on developing methods to mitigate this limitation, employing architectures like recurrent neural networks, Kalman filters, and transformers, along with algorithms that incorporate uncertainty representation and efficient planning techniques under partial observation. These advancements are crucial for improving the performance and robustness of AI systems in complex real-world applications, such as autonomous driving, robotics, and traffic control, where complete information is rarely available. The ultimate goal is to enable agents to make optimal decisions despite incomplete knowledge of their surroundings.

Papers

February 19, 2024

Graph-based Virtual Sensing from Sparse and Partial Multivariate Observations
Giovanni De Felice, Andrea Cini, Daniele Zambon, Vladimir V. Gusev, Cesare Alippi
Many Sparse Partial Observability Virtual Sensor Multivariate Model Sensory Coverage

February 8, 2024

Offline Risk-sensitive RL with Partial Observability to Enhance Performance in Human-Robot Teaming
Giorgio Angelotti, Caroline P. C. Chanel, Adam H. M. Pinto, Christophe Lounis, Corentin Chauffaut, Nicolas Drougard
System Performance Physiological Signal Observable Markov Decision Process Partial Observability Human Robot Teaming Human Operator Offline Safe Reinforcement Learning Human State Estimation

December 20, 2023

Collaborative Optimization of the Age of Information under Partial Observability
Anam Tahir, Kai Cui, Bastian Alt, Amr Rizk, Heinz Koeppl
Full Information Speech Based Age Partial Observability Robotic Agent Agent Communication Channel Wise Processing Pipeline Collaborative Optimization

December 18, 2023

December 10, 2023

Learning the Causal Structure of Networked Dynamical Systems under Latent Nodes and Structured Noise
Augusto Santos, Diogo Rente, Rui Seabra, José M. F. Moura
LeArning Abstract Causal Inference Dynamical System Latent Representation Causal Structure Partial Observability Correlated Noise Causal Network

December 6, 2023

Weathering Ongoing Uncertainty: Learning and Planning in a Time-Varying Partially Observable Environment
Gokul Puthumanaillam, Xiangyu Liu, Negar Mehr, Melkior Ornik
LeArning Abstract Markov Decision Process Task Planning State Estimation Partial Observability Observable Environment Uncertain Dynamic Environment Time Varying Environment

November 20, 2023

Provable Representation with Efficient Planning for Partial Observable Reinforcement Learning
Hongming Zhang, Tongzheng Ren, Chenjun Xiao, Dale Schuurmans, Bo Dai
Markov Decision Process Partial Observability Partially Observable Markov Decision Process Efficient Planning Reliable Representation

November 7, 2023

Multi-View Causal Representation Learning with Partial Observability
Dingling Yao, Danru Xu, Sébastien Lachapelle, Sara Magliacane, Perouz Taslakian, Georg Martius, Julius von Kügelgen, Francesco Locatello
Latent Variable Causal Representation Learning Partial Observability Nonlinear ICA

September 30, 2023

Making Friends in the Dark: Ad Hoc Teamwork Under Partial Observability
João G. Ribeiroa, Cassandro Martinhoa, Alberto Sardinhaa, Francisco S. Melo
Prior Knowledge Partial Observability First Principle C POMDPs Unknown Teammate

September 29, 2023

Adversarial Imitation Learning from Visual Observations using Latent Information
Vittorio Giammarino, James Queeney, Ioannis Ch. Paschalidis
Reinforcement Learning High Dimensional Partial Observability Adversarial Imitation Learning Visual Exploration Latent Topic Policy Adversarial Imitation Learning

September 25, 2023

Scene Informer: Anchor-based Occlusion Inference and Trajectory Prediction in Partially Observable Environments
Bernard Lange, Jiachen Li, Mykel J. Kochenderfer
Trajectory Prediction Occupancy Prediction Partial Observability Observable Environment Agent Trajectory Occlusion Sensitivity Analysis Occlusion Inference Scene Knowledge

September 9, 2023

Verifiable Reinforcement Learning Systems via Compositionality
Cyrus Neary, Aryaman Singh Samyal, Christos Verginis, Murat Cubuktepe, Ufuk Topcu
Action Space Partial Observability Compositional Language Deep Reinforcement Learning Agent Compositional Reinforcement Learning

June 22, 2023

DiMSam: Diffusion Models as Samplers for Task and Motion Planning under Partial Observability
Xiaolin Fang, Caelan Reed Garrett, Clemens Eppner, Tomás Lozano-Pérez, Leslie Pack Kaelbling, Dieter Fox
Diffusion Model Motion Planning Long Term Related Task Deep Generative Model Partial Observability Long Horizon Generative Learning Proximal Sampler

June 19, 2023

Learning Models of Adversarial Agent Behavior under Partial Observability
Sean Ye, Manisha Natarajan, Zixuan Wu, Rohan Paleja, Letian Chen, Matthew C. Gombolay
Adversarial Attack Adversarial Learning Partial Observability Adversarial Behavior Learning Model Adversarial Agent

June 1, 2023

Knowledge-based Reasoning and Learning under Partial Observability in Ad Hoc Teamwork
Hasra Dodampegama, Mohan Sridharan
LeArning Abstract Partial Observability Prior Coordination Knowledge Reasoning Ad Hoc Agent Benchmark Team Formation Prior Commonsense Domain Knowledge

May 22, 2023

Bandit Submodular Maximization for Multi-Robot Coordination in Unpredictable and Partially Observable Environments
Zirui Xu, Xiaofeng Lin, Vasileios Tzoumas
Submodular Maximization Partial Observability Predictability Problem Multi Agent Coordination Observable Environment Optimal Action Multi Robot Coordination Submodular Bandit

May 17, 2023

Assessing the Impact of Context Inference Error and Partial Observability on RL Methods for Just-In-Time Adaptive Interventions
Karine Karine, Predrag Klasnja, Susan A. Murphy, Benjamin M. Marlin
Reinforcement Learning Global Impact Policy Gradient Partial Observability Adaptive Intervention Intervention Policy RL Method

March 10, 2023

Decision-Making Under Uncertainty: Beyond Probabilities
Thom Badings, Thiago D. Simão, Marnix Suilen, Nils Jansen
Reinforcement Learning Markov Decision Process High Uncertainty Anticipation Epistemic Uncertainty Aleatoric Uncertainty Partial Observability Relative Probability

March 3, 2023

Data Association Aware POMDP Planning with Hypothesis Pruning Performance Guarantees
Moran Barenboim, Idan Lev-Yehudi, Vadim Indelman
Markov Decision Process Partial Observability Based Pruning POMDP Planning POMDP Solver

Partial Observability

Papers

Graph-based Virtual Sensing from Sparse and Partial Multivariate Observations

Offline Risk-sensitive RL with Partial Observability to Enhance Performance in Human-Robot Teaming

Collaborative Optimization of the Age of Information under Partial Observability

Inferring the Graph of Networked Dynamical Systems under Partial Observability and Spatially Colored Noise

Robust Active Measuring under Model Uncertainty

Learning the Causal Structure of Networked Dynamical Systems under Latent Nodes and Structured Noise

Weathering Ongoing Uncertainty: Learning and Planning in a Time-Varying Partially Observable Environment

Provable Representation with Efficient Planning for Partial Observable Reinforcement Learning

Multi-View Causal Representation Learning with Partial Observability

Making Friends in the Dark: Ad Hoc Teamwork Under Partial Observability

Adversarial Imitation Learning from Visual Observations using Latent Information

Scene Informer: Anchor-based Occlusion Inference and Trajectory Prediction in Partially Observable Environments

Verifiable Reinforcement Learning Systems via Compositionality

DiMSam: Diffusion Models as Samplers for Task and Motion Planning under Partial Observability

Learning Models of Adversarial Agent Behavior under Partial Observability

Knowledge-based Reasoning and Learning under Partial Observability in Ad Hoc Teamwork

Bandit Submodular Maximization for Multi-Robot Coordination in Unpredictable and Partially Observable Environments

Assessing the Impact of Context Inference Error and Partial Observability on RL Methods for Just-In-Time Adaptive Interventions

Decision-Making Under Uncertainty: Beyond Probabilities

Data Association Aware POMDP Planning with Hypothesis Pruning Performance Guarantees