CraftAssist: A Framework for Dialogue-enabled Interactive Agents
arXiv:1907.08584
Abstract
This paper describes an implementation of a bot assistant in Minecraft, and the tools and platform allowing players to interact with the bot and to record those interactions. The purpose of building such an assistant is to facilitate the study of agents that can complete tasks specified by dialogue, and eventually, to learn from dialogue interactions.
References in corpus (12)
- Playing Atari with Deep Reinforcement Learning
- VQA: Visual Question Answering
- SQuAD: 100,000+ Questions for Machine Comprehension of Text
- Seq2SQL: Generating Structured Queries from Natural Language using Reinforcement Learning
- AI2-THOR: An Interactive 3D Environment for Visual AI
- Control of Memory, Active Perception, and Action in Minecraft
- Language to Logical Form with Neural Attention
- Hierarchical and Interpretable Skill Acquisition in Multi-task Reinforcement Learning
- The MineRL 2019 Competition on Sample Efficient Reinforcement Learning using Human Priors
- Data Recombination for Neural Semantic Parsing
- Deep Reinforcement Learning with Model Learning and Monte Carlo Tree Search in Minecraft
- CraftAssist Instruction Parsing: Semantic Parsing for a Minecraft Assistant
Cited by in corpus (8)
- On the Opportunities and Risks of Foundation Models
- Why Build an Assistant in Minecraft?
- The MineRL BASALT Competition on Learning from Human Feedback
- Learning Adaptive Language Interfaces through Decomposition
- Neural Abstructions: Abstractions that Support Construction for Grounded Language Learning
- Mind Reading at Work: Cooperation without common ground
- Using reinforcement learning to design an AI assistantfor a satisfying co-op experience
- Feudal Reinforcement Learning by Reading Manuals