Grounding Natural Language Commands to StarCraft II Game States for Narration-Guided Reinforcement Learning

04/24/2019
by   Nicholas Waytowich, et al.
0

While deep reinforcement learning techniques have led to agents that are successfully able to learn to perform a number of tasks that had been previously unlearnable, these techniques are still susceptible to the longstanding problem of reward sparsity. This is especially true for tasks such as training an agent to play StarCraft II, a real-time strategy game where reward is only given at the end of a game which is usually very long. While this problem can be addressed through reward shaping, such approaches typically require a human expert with specialized knowledge. Inspired by the vision of enabling reward shaping through the more-accessible paradigm of natural-language narration, we investigate to what extent we can contextualize these narrations by grounding them to the goal-specific states. We present a mutual-embedding model using a multi-input deep-neural network that projects a sequence of natural language commands into the same high-dimensional representation space as corresponding goal states. We show that using this model we can learn an embedding space with separable and distinct clusters that accurately maps natural-language commands to corresponding game states . We also discuss how this model can allow for the use of narrations as a robust form of reward shaping to improve RL performance and efficiency.

READ FULL TEXT
research
10/31/2019

A Narration-based Reward Shaping Approach using Grounded Natural Language Commands

While deep reinforcement learning techniques have led to agents that are...
research
02/20/2019

From Language to Goals: Inverse Reinforcement Learning for Vision-Based Instruction Following

Reinforcement learning is a promising framework for solving control prob...
research
12/02/2017

Interactive Reinforcement Learning for Object Grounding via Self-Talking

Humans are able to identify a referred visual object in a complex scene ...
research
11/06/2018

An Optimal Itinerary Generation in a Configuration Space of Large Intellectual Agent Groups with Linear Logic

A group of intelligent agents which fulfill a set of tasks in parallel i...
research
09/23/2021

Reinforced Natural Language Interfaces via Entropy Decomposition

In this paper, we study the technical problem of developing conversation...
research
11/03/2022

lilGym: Natural Language Visual Reasoning with Reinforcement Learning

We present lilGym, a new benchmark for language-conditioned reinforcemen...
research
05/08/2017

The Pragmatics of Indirect Commands in Collaborative Discourse

Today's artificial assistants are typically prompted to perform tasks th...

Please sign up or login with your details

Forgot password? Click here to reset