Counting to Explore and Generalize in Text-based Games
arXiv:1806.11525
Abstract
We propose a recurrent RL agent with an episodic exploration mechanism that helps discovering good policies in text-based game environments. We show promising results on a set of generated text-based games of varying difficulty where the goal is to collect a coin located at the end of a chain of rooms. In contrast to previous text-based RL approaches, we observe that our agent learns policies that generalize to unseen games of greater difficulty.
References in corpus (7)
- Adam: A Method for Stochastic Optimization
- Deep Exploration via Bootstrapped DQN
- Parameter Space Noise for Exploration
- Count-Based Exploration with Neural Density Models
- Language Understanding for Text-based Games Using Deep Reinforcement Learning
- Deep Reinforcement Learning with a Natural Language Action Space
- What can you do with a rock? Affordance extraction via word embeddings
Cited by in corpus (17)
- Learn What Not to Learn: Action Elimination with Deep Reinforcement Learning
- Learning Dynamic Belief Graphs to Generalize on Text-Based Games
- ALFWorld: Aligning Text and Embodied Environments for Interactive Learning
- NAIL: A General Interactive Fiction Agent
- How to Avoid Being Eaten by a Grue: Structured Exploration Strategies for Textual Worlds
- Interactive Fiction Games: A Colossal Adventure
- Deep Reinforcement Learning with Stacked Hierarchical Attention for Text-based Games
- Action Assembly: Sparse Imitation Learning for Text Based Games with Combinatorial Action Spaces
- Towards Socially Intelligent Agents with Mental State Transition and Human Utility
- Zero-Shot Learning of Text Adventure Games with Sentence-Level Semantics
- An Overview of Natural Language State Representation for Reinforcement Learning
- Interactive Machine Comprehension with Dynamic Knowledge Graphs
- Bootstrapped Q-learning with Context Relevant Observation Pruning to Generalize in Text-based Games
- Neuro-Symbolic Reinforcement Learning with First-Order Logic
- Towards Goal-Oriented Agents for Evolving Problems Observed via Conversation
- Learning to Generalize for Sequential Decision Making
- LOA: Logical Optimal Actions for Text-based Interaction Games