Neural Architecture Search of SPD Manifold Networks
This work addresses the challenge of designing efficient SPD networks for applications such as recognition tasks, though it is incremental as it extends NAS to a specific domain.
The paper tackles the problem of automating the design of Symmetric Positive Definite (SPD) manifold networks by proposing a neural architecture search (NAS) method, resulting in models that achieve better performance on tasks like drone, action, and emotion recognition and are over three times lighter than those from state-of-the-art NAS algorithms.
In this paper, we propose a new neural architecture search (NAS) problem of Symmetric Positive Definite (SPD) manifold networks, aiming to automate the design of SPD neural architectures. To address this problem, we first introduce a geometrically rich and diverse SPD neural architecture search space for an efficient SPD cell design. Further, we model our new NAS problem with a one-shot training process of a single supernet. Based on the supernet modeling, we exploit a differentiable NAS algorithm on our relaxed continuous search space for SPD neural architecture search. Statistical evaluation of our method on drone, action, and emotion recognition tasks mostly provides better results than the state-of-the-art SPD networks and traditional NAS algorithms. Empirical results show that our algorithm excels in discovering better performing SPD network design and provides models that are more than three times lighter than searched by the state-of-the-art NAS algorithms.