RSS feed Add to your Facebook page LinkedIn Watch us on Youtube

Unsupervised Machine Learning in the Military Domain

IST-ET-113 Exploratory Team (ET) on Unsupervised Machine Learning in the Military Domain, 27 May 2021, virtually

Credit: Flickr

Several military applications that require the use of Artificial Intelligence (AI) suffer from having a limited amount of labelled data (i.e. real data acquired from the battlefield), which is necessary for training existing deep learning supervised models. The availability of new approaches that can reach appropriate intelligent solutions without the need of labelled data would be of great use in several military domains (situation awareness, autonomous vehicle driving, analysis of big military data, etc.). IST-ET-113 aims to extend the supervised existing solutions in such a way that they can work in an unsupervised setting too, by starting from existing neural models (e.g., neural trees, CNN, etc.) and trying to adapt them to work in an unsupervised way. This Exploratory Team intends to establish innovative solutions to

  1. Develop new deep learning architectures able to work in an unsupervised way;
  2. Develop new deep learning algorithms able to work with limited amount of annotated data;
  3. Develop intelligent transfer learning methods for using available (also civil) data sets for military applications;
  4. Reduce the costs for the provision of assets (e.g. data set annotation, QS, …); and
  5. Acquire new techniques for incremental learning in military applications.

This activity will contribute to NATO’s vision on Unsupervised Machine Learning in the field of military applications. IST-ET-113 held its latest meeting, virtually, on 27 May 2021. For further information, please contact the IST Panel Office.​​

Published by IST