GOLD-NAS: Gradual, One-Level, Differentiable
This work addresses the need for more flexible and efficient neural architecture search methods in machine learning, though it appears incremental as it builds on existing differentiable NAS approaches.
The paper tackles the problem of constrained search flexibility in neural architecture search by relaxing manual constraints to create a vast search space, and proposes GOLD-NAS, a novel algorithm that gradually prunes weak operators to find Pareto-optimal architectures with a tradeoff between accuracy and model complexity.
There has been a large literature of neural architecture search, but most existing work made use of heuristic rules that largely constrained the search flexibility. In this paper, we first relax these manually designed constraints and enlarge the search space to contain more than $10^{160}$ candidates. In the new space, most existing differentiable search methods can fail dramatically. We then propose a novel algorithm named Gradual One-Level Differentiable Neural Architecture Search (GOLD-NAS) which introduces a variable resource constraint to one-level optimization so that the weak operators are gradually pruned out from the super-network. In standard image classification benchmarks, GOLD-NAS can find a series of Pareto-optimal architectures within a single search procedure. Most of the discovered architectures were never studied before, yet they achieve a nice tradeoff between recognition accuracy and model complexity. We believe the new space and search algorithm can advance the search of differentiable NAS.