Efficient Model Performance Estimation via Feature Histories

TL;DR

This paper introduces a flexible early-performance estimation method for neural networks using feature histories, improving hyper-parameter optimization and neural architecture search efficiency without restrictive search space constraints.

Contribution

The proposed method leverages feature evolution during early training to accurately estimate maximum performance, enhancing search efficiency in HPO and NAS.

Findings

Achieves better architectures than DARTS.

Reduces search time by 80%.

Compatible with multiple search algorithms.

Abstract

An important step in the task of neural network design, such as hyper-parameter optimization (HPO) or neural architecture search (NAS), is the evaluation of a candidate model's performance. Given fixed computational resources, one can either invest more time training each model to obtain more accurate estimates of final performance, or spend more time exploring a greater variety of models in the configuration space. In this work, we aim to optimize this exploration-exploitation trade-off in the context of HPO and NAS for image classification by accurately approximating a model's maximal performance early in the training process. In contrast to recent accelerated NAS methods customized for certain search spaces, e.g., requiring the search space to be differentiable, our method is flexible and imposes almost no constraints on the search space. Our method uses the evolution history of features of a network during the early stages of training to build a proxy classifier that matches the peak performance of the network under consideration. We show that our method can be combined with multiple search algorithms to find better solutions to a wide range of tasks in HPO and NAS. Using a sampling-based search algorithm and parallel computing, our method can find an architecture which is better than DARTS and with an 80% reduction in wall-clock search time.