D-DARTS: Distributed Differentiable Architecture Search
This work addresses a bottleneck in neural architecture search for computer vision researchers, offering an incremental improvement over DARTS.
The paper tackles the problem of DARTS reducing search space and excluding promising architectures by proposing D-DARTS, which uses nesting at the cell level instead of weight-sharing to produce more diversified architectures, achieving competitive performance on multiple computer vision tasks.
Differentiable ARchiTecture Search (DARTS) is one of the most trending Neural Architecture Search (NAS) methods. It drastically reduces search cost by resorting to weight-sharing. However, it also dramatically reduces the search space, thus excluding potential promising architectures. In this article, we propose D-DARTS, a solution that addresses this problem by nesting neural networks at the cell level instead of using weight-sharing to produce more diversified and specialized architectures. Moreover, we introduce a novel algorithm that can derive deeper architectures from a few trained cells, increasing performance and saving computation time. In addition, we also present an alternative search space (DARTOpti) in which we optimize existing handcrafted architectures (e.g., ResNet) rather than starting from scratch. This approach is accompanied by a novel metric that measures the distance between architectures inside our custom search space. Our solution reaches competitive performance on multiple computer vision tasks. Code and pretrained models can be accessed at https://github.com/aheuillet/D-DARTS.