IGLU Gridworld: Simple and Fast Environment for Embodied Dialog Agents

05/31/2022

∙

We present the IGLU Gridworld: a reinforcement learning environment for building and evaluating language conditioned embodied agents in a scalable way. The environment features visual agent embodiment, interactive learning through collaboration, language conditioned RL, and combinatorically hard task (3d blocks building) space.

READ FULL TEXT

IGLU Gridworld: Simple and Fast Environment for Embodied Dialog Agents

Simple Embodied Language Learning as a Byproduct of Meta-Reinforcement Learning

VALAN: Vision and Language Agent Navigation

LAD: Language Augmented Diffusion for Reinforcement Learning

VisualHints: A Visual-Lingual Environment for Multimodal Reinforcement Learning

EST: Evaluating Scientific Thinking in Artificial Agents

Flatland: a Lightweight First-Person 2-D Environment for Reinforcement Learning

Inverse reinforcement learning conditioned on brain scan

IGLU Gridworld: Simple and Fast Environment for Embodied Dialog Agents

Related Research

Simple Embodied Language Learning as a Byproduct of Meta-Reinforcement Learning

VALAN: Vision and Language Agent Navigation

LAD: Language Augmented Diffusion for Reinforcement Learning

VisualHints: A Visual-Lingual Environment for Multimodal Reinforcement Learning

EST: Evaluating Scientific Thinking in Artificial Agents

Flatland: a Lightweight First-Person 2-D Environment for Reinforcement Learning

Inverse reinforcement learning conditioned on brain scan