Training a reinforcement learning agent to carry out natural language instructions is limited by the available supervision, i.e. knowing when the instruction has been carried out. We adapt the CLEVR visual question answering dataset to generate complex natural language navigation instructions and accompanying scene graphs, yielding an environment-agnostic supervised dataset. To demonstrate the use of this data set, we map the scenes to the VizDoom environment and use the architecture in \citet{gatedattention} to train an agent to carry out these more complex language instructions.
@article{arxiv.2106.01607,
title = {Grounding Complex Navigational Instructions Using Scene Graphs},
author = {Michiel de Jong and Satyapriya Krishna and Anuva Agarwal},
journal= {arXiv preprint arXiv:2106.01607},
year = {2021}
}
Comments
arXiv admin note: text overlap with arXiv:1706.07230 by other authors