We anticipate that the use of AI for future tactical systems and applications will only increase over time.
Our core AI technology, Surprise Based Learning (SBL), gives autonomous platforms the ability to learn and plan in an unknown environment without any prior knowledge of its actions or their impact on the environment. When combined with deep learning (DL) and reinforcement learning (RL), SBL profoundly improves their benefits for a broad range of tactical applications. SBL is capable of detecting unforeseen or anomalous trends that were not learned by DL or RL during its training phase, and adapting to these trends without un-learning / re-learning.
On their own, both DL and RL requires massive computing resources (e.g. supercomputer or large amount of cloud computing) whenever a new trend is observed or new data is obtained to update their original system parameters. Un-learning / re-learning requires both the old and new datasets, and thus massive computing resources. Therefore, SBL minimizes or even eliminates the need for the resources originally required for un-learning / re-learning. This makes our TACTICAL AI (the combination of SBL, DL, and RL) directly applicable to tactical missions when autonomous platforms cannot afford to un-learn / re-learn in real time.