Searching Efficient Deep Architectures for Radar Target Detection using Monte-Carlo Tree Search
This work addresses the challenge of deploying deep learning in embedded radar systems by reducing computational complexity, though it is incremental as it applies an existing NAS method to a specific domain.
The paper tackled the problem of high computational complexity in deep neural networks for radar target detection by using Monte-Carlo Tree Search-based neural architecture search to find efficient architectures. It resulted in a novel network that meets detection probability requirements while being significantly lighter than a baseline, with specific performance metrics evaluated on endoclutter radar signals.
Recent research works establish deep neural networks as high performing tools for radar target detection, especially on challenging environments (presence of clutter or interferences, multi-target scenarii...). However, the usually large computational complexity of these networks is one of the factors preventing them from being widely implemented in embedded radar systems. We propose to investigate novel neural architecture search (NAS) methods, based on Monte-Carlo Tree Search (MCTS), for finding neural networks achieving the required detection performance and striving towards a lower computational complexity. We evaluate the searched architectures on endoclutter radar signals, in order to compare their respective performance metrics and generalization properties. A novel network satisfying the required detection probability while being significantly lighter than the expert-designed baseline is proposed.