Maze Environment

Maze environments serve as a versatile benchmark for evaluating various AI algorithms, particularly in navigation, planning, and reinforcement learning. Current research focuses on improving the efficiency and robustness of these algorithms, exploring architectures like transformers and recurrent neural networks, and investigating the impact of factors such as goal specification, memory limitations, and multi-agent coordination. This research is significant because it advances our understanding of AI capabilities in complex scenarios and has implications for real-world applications such as robotics, autonomous navigation, and explainable AI.

Papers

September 18, 2023

Learning Spatial and Temporal Hierarchies: Hierarchical Active Inference for navigation in Multi-Room Maze Environments
Daria de Tinguy, Toon Van de Maele, Tim Verbelen, Bart Dhoedt
Local Navigation Cognitive Map Maze Environment Spatial Learning Temporal Structure Curiosity Driven Exploration Egocentric World Model

August 1, 2023

Target Search and Navigation in Heterogeneous Robot Systems with Deep Reinforcement Learning
Yun Chen, Jiaping Xiao
Reinforcement Learning Deep Reinforcement Learning Local Navigation Heterogeneous Robot Maze Environment

June 23, 2023

Inferring Hierarchical Structure in Multi-Room Maze Environments
Daria de Tinguy, Toon Van de Maele, Tim Verbelen, Bart Dhoedt
Cognitive Map Hierarchical Structure Maze Environment Curiosity Driven Exploration Egocentric World Model

June 1, 2023

Investigating Navigation Strategies in the Morris Water Maze through Deep Reinforcement Learning
Andrew Liu, Alla Borisyuk
Deep Reinforcement Learning Internal Representation Deep Reinforcement Artificial Agent Maze Environment Navigation Strategy

March 2, 2023

Navigating the Metric Maze: A Taxonomy of Evaluation Metrics for Anomaly Detection in Time Series
Sondre Sørbø, Massimiliano Ruocco
Anomaly Detection Time Series Comprehensive Taxonomy Evaluation Metric Metric Library Time Series Anomaly Detection Maze Environment

November 1, 2022

Discrete Factorial Representations as an Abstraction for Goal Conditioned Reinforcement Learning
Riashat Islam, Hongyu Zang, Anirudh Goyal, Alex Lamb, Kenji Kawaguchi, Xin Li, Romain Laroche, Yoshua Bengio, Remi Tachet Des Combes
Cognitive Abstraction Goal Conditioned Reinforcement Learning Discrete Representation Maze Environment Goal Conditioned Agent

October 31, 2022

Disentangled (Un)Controllable Features
Jacob E. Kooi, Mark Hoogendoorn, Vincent François-Lavet
High Dimensional Latent Feature Low Dimensional Representation Maze Environment Disentangled Haze

October 24, 2022

Evaluating Long-Term Memory in 3D Mazes
Jurgis Pasukonis, Timothy Lillicrap, Danijar Hafner
Reinforcement Learning Memory Trace Long Term Memory Reinforcement Learning Benchmark Maze Environment Maze Navigation

September 18, 2022

A Computational Model of Learning Flexible Navigation in a Maze by Layout-Conforming Replay of Place Cells
Yuanxiang Gao
Computational Model Maze Environment RePLAy Loss Replay Based Feasible Reward Place Cell Trajectory Replay

September 12, 2022

A Differentiable Loss Function for Learning Heuristics in A*
Leah Chrestien, Tomas Pevny, Antonin Komenda, Stefan Edelkamp
Deep Neural Network Optimization Purpose State of the Art Deep Maze Environment Heuristic Function Heuristic Learning Differentiable Loss

June 9, 2022

There is no Accuracy-Interpretability Tradeoff in Reinforcement Learning for Mazes
Yishay Mansour, Michal Moshkovitz, Cynthia Rudin
Reinforcement Learning Inherent Interpretability Maze Environment Reinforcement Learning System

May 11, 2022

Developing cooperative policies for multi-stage reinforcement learning tasks
Jordan Erskine, Chris Lehnert
Reinforcement Learning Hierarchical Reinforcement Learning Long Horizon Task Maze Environment Sequential Decision Making Policy

January 16, 2022

Standby-Based Deadlock Avoidance Method for Multi-Agent Pickup and Delivery Tasks
Tomoki Yamauchi, Yuki Miyashita, Toshiharu Sugawara
Multiple Agent Maze Environment Multi Agent Pickup Delivery Related Task

November 29, 2021

Improving Experience Replay with Successor Representation
Yizhi Yuan, Marcelo G Mattar
Reinforcement Learning Experience Replay Maze Environment RePLAy Loss Successor Representation

November 17, 2021

Search and Rescue in a Maze-like Environment with Ant and Dijkstra Algorithms
Z. Husain, A. Al Zaabi, H. Hildmann, F. Saffre, D. Ruta, A. F. Isakovic
Search Query Autonomous Agent Search and Rescue Maze Environment Ant Colony Dijkstra Algorithm Autonomous Mobile Agent Ant Nesting Algorithm Agent Swarm

Maze Environment

Papers

Learning Spatial and Temporal Hierarchies: Hierarchical Active Inference for navigation in Multi-Room Maze Environments

Target Search and Navigation in Heterogeneous Robot Systems with Deep Reinforcement Learning

Inferring Hierarchical Structure in Multi-Room Maze Environments

Investigating Navigation Strategies in the Morris Water Maze through Deep Reinforcement Learning

Navigating the Metric Maze: A Taxonomy of Evaluation Metrics for Anomaly Detection in Time Series

Discrete Factorial Representations as an Abstraction for Goal Conditioned Reinforcement Learning

Disentangled (Un)Controllable Features

Evaluating Long-Term Memory in 3D Mazes

A Computational Model of Learning Flexible Navigation in a Maze by Layout-Conforming Replay of Place Cells

A Differentiable Loss Function for Learning Heuristics in A*

There is no Accuracy-Interpretability Tradeoff in Reinforcement Learning for Mazes

Developing cooperative policies for multi-stage reinforcement learning tasks

Standby-Based Deadlock Avoidance Method for Multi-Agent Pickup and Delivery Tasks

Improving Experience Replay with Successor Representation

Search and Rescue in a Maze-like Environment with Ant and Dijkstra Algorithms