Learning State Representations in Complex Systems with Multimodal Data
Authors:
Pavel Solovev,
Vladimir Aliev,
Pavel Ostyakov,
Gleb Sterkin,
Elizaveta Logacheva,
Stepan Troeshestov,
Roman Suvorov,
Anton Mashikhin,
Oleg Khomenko,
Sergey I. Nikolenko
Abstract:
Representation learning becomes especially important for complex systems with multimodal data sources such as cameras or sensors. Recent advances in reinforcement learning and optimal control make it possible to design control algorithms on these latent representations, but the field still lacks a large-scale standard dataset for unified comparison. In this work, we present a large-scale dataset a…
▽ More
Representation learning becomes especially important for complex systems with multimodal data sources such as cameras or sensors. Recent advances in reinforcement learning and optimal control make it possible to design control algorithms on these latent representations, but the field still lacks a large-scale standard dataset for unified comparison. In this work, we present a large-scale dataset and evaluation framework for representation learning for the complex task of landing an airplane. We implement and compare several approaches to representation learning on this dataset in terms of the quality of simple supervised learning tasks and disentanglement scores. The resulting representations can be used for further tasks such as anomaly detection, optimal control, model-based reinforcement learning, and other applications.
△ Less
Submitted 15 January, 2019; v1 submitted 27 November, 2018;
originally announced November 2018.
