1. An Active Learning Framework for Constructing High-Fidelity Mobility Maps.
- Author
-
Marple, Gary R., Gorsich, David, Jayakumar, Paramsothy, and Veerapaneni, Shravan
- Subjects
- *
ACTIVE learning , *MACHINE learning , *STATISTICAL sampling , *TRAINING needs , *ALGORITHMS - Abstract
Recent workat the U.S. Army CCDC Ground Vehicle Systems Center has shown that machine learning classifiers can quickly construct high-fidelity mobility maps. Training these classifiers, on the other hand, is still a challenge, since each data instance is labeled by performing a computationally intensive, physics-based simulation. In this paper we introduce an active learning framework, based on the query-by-bagging algorithm, that substantially reduces the number of simulations needed to train a classifier. Experimental results suggest that our sampling algorithm can train a neural network, with higher accuracy, using less than half the number of simulations when compared to random sampling. [ABSTRACT FROM AUTHOR]
- Published
- 2021
- Full Text
- View/download PDF