Efficient Automation of Neural Network Design: A Survey on Differentiable Neural Architecture Search
It provides a comprehensive overview and taxonomy for researchers in automated machine learning, but is incremental as it synthesizes existing work.
This survey reviews Differentiable Neural Architecture Search (DNAS), which automates neural network design and is faster by several orders of magnitude with fewer computational resources compared to previous methods like Reinforcement Learning or Evolutionary Algorithms.
In the past few years, Differentiable Neural Architecture Search (DNAS) rapidly imposed itself as the trending approach to automate the discovery of deep neural network architectures. This rise is mainly due to the popularity of DARTS, one of the first major DNAS methods. In contrast with previous works based on Reinforcement Learning or Evolutionary Algorithms, DNAS is faster by several orders of magnitude and uses fewer computational resources. In this comprehensive survey, we focus specifically on DNAS and review recent approaches in this field. Furthermore, we propose a novel challenge-based taxonomy to classify DNAS methods. We also discuss the contributions brought to DNAS in the past few years and its impact on the global NAS field. Finally, we conclude by giving some insights into future research directions for the DNAS field.