Understanding Patterns of Deep Learning ModelEvolution in Network Architecture Search

TL;DR

This paper analyzes how deep learning models evolve over time during network architecture search, revealing patterns influenced by the search algorithm and implications for caching, scheduling, and model popularity.

Contribution

It provides a quantitative analysis of model evolution patterns in NAS, highlighting the influence of regularized evolution and distributed settings on architecture dynamics.

Findings

Evolution influenced by regularized search algorithms

Distributed settings exhibit distinct evolutionary patterns

Model popularity depends on donor frequency in sliding windows

Abstract

Network Architecture Search and specifically Regularized Evolution is a common way to refine the structure of a deep learning model.However, little is known about how models empirically evolve over time which has design implications for designing caching policies, refining the search algorithm for particular applications, and other important use cases.In this work, we algorithmically analyze and quantitatively characterize the patterns of model evolution for a set of models from the Candle project and the Nasbench-201 search space.We show how the evolution of the model structure is influenced by the regularized evolution algorithm. We describe how evolutionary patterns appear in distributed settings and opportunities for caching and improved scheduling. Lastly, we describe the conditions that affect when particular model architectures rise and fall in popularity based on their frequency of acting as a donor in a sliding window.