Causal Exploration

Causal exploration aims to improve the efficiency and reliability of machine learning by leveraging causal relationships within data. Current research focuses on developing algorithms that can identify and utilize these relationships for tasks like reinforcement learning and wind estimation, often employing techniques like attention mechanisms and causal discovery to guide exploration and model training. This approach promises to enhance sample efficiency in various applications, leading to more robust and data-efficient models, particularly in scenarios with limited data or noisy observations. The resulting improvements in model accuracy and interpretability are significant for both scientific understanding and practical deployment of AI systems.

Papers

July 30, 2024

Boosting Efficiency in Task-Agnostic Exploration through Causal Knowledge
Yupei Yang, Biwei Huang, Shikui Tu, Lei Xu
Causal Knowledge Enhancing Efficiency Causal Insight Agnostic Exploration Agnostic Reinforcement Learning Causal Exploration

July 17, 2024

Variable-Agnostic Causal Exploration for Reinforcement Learning
Minh Hoang Nguyen, Hung Le, Svetha Venkatesh
Reinforcement Learning Causal Discovery Modern Reinforcement Learning Agent Exploration Causal Exploration

July 1, 2024

Wind Estimation in Unmanned Aerial Vehicles with Causal Machine Learning
Abdulaziz Alwalan, Miguel Arana-Catania
Unmanned Aerial Vehicle Optimal Trajectory Multi UAV Causal Machine Learning Resolution Wind Causal Exploration

May 15, 2024

Advancing Explainable AI with Causal Analysis in Large-Scale Fuzzy Cognitive Maps
Marios Tyrovolas, Nikolaos D. Kallimanis, Chrysostomos Stylios
Explainable AI Causal Effect Causal Analysis Fuzzy Cognitive Map Causal Exploration

Causal Exploration

Papers

Boosting Efficiency in Task-Agnostic Exploration through Causal Knowledge

Variable-Agnostic Causal Exploration for Reinforcement Learning

Wind Estimation in Unmanned Aerial Vehicles with Causal Machine Learning

Advancing Explainable AI with Causal Analysis in Large-Scale Fuzzy Cognitive Maps