27 datasets found

A new simulation environment designed as a simple testbed for demonstrating the utility of RL as a tool for concept analysis with military applications as well as to aid with...

The dataset used in the paper is a hierarchical structure of propositions, where a higher-level proposition is a temporal abstraction of lower-level propositions. Each...

The dataset used in the paper is a multi-agent reinforcement learning dataset, where agents learn to coordinate their actions to achieve a common goal.

Offline Fictitious Self-Play for Competitive Games

The authors used the Coin Game, a social dilemma game with visual observations, to evaluate their approach.

The dataset used in the paper is the SMAC dataset, a multi-agent reinforcement learning environment.

The dataset used in the paper is a stochastic game, specifically a simultaneous-move game, with multiple agents and a finite horizon.

The dataset used in the paper is a collection of 21 diverse multi-agent tasks in four environments: level-based foraging, boulder-push, multi-robot warehouse, and multi-agent...

Pommerman is a 2-vs-2 game environment introduced in 2018, and is a modified version of the popular game 'Bomberman'. It is a good testbed for RL, because the environment...

Multi-Agent Particle Environment (MAPE) game environment

Hierarchical RNNs-Based Transformers MADDPG for Mixed Cooperative-Competitive Environments

Packet routing environments for cooperative multi-agent reinforcement learning

Packet routing environments for cooperative multi-agent reinforcement learning

Packet routing environments for cooperative multi-agent reinforcement learning

Packet routing environments for cooperative multi-agent reinforcement learning

The Half Field Offense environment is a multi-agent reinforcement learning problem where multiple agents interact in a soccer-like environment.

The Multi-Team Predator-Prey environment is a multi-agent reinforcement learning problem where multiple predators and prey interact in a grid-based environment.

The authors used a multi-agent grid-world game called COLLECT, where four agents can move and collect coins of different colors.

The proposed solution is divided into two main components: leverage human-MARL teaming framework, called COGMENT, to handle the collaboration between human and multiple RL-based...

The dataset used in the paper is the StarCraft II micromanagement tasks.