AdaXpert: Adapting Neural Architecture for Growing Data

TL;DR

AdaXpert is a method that dynamically adjusts neural network architectures to handle growing data volumes and classes efficiently, ensuring promising performance without unnecessary adjustments.

Contribution

It introduces an architecture adjuster and an adaptation condition to promptly and efficiently adapt models to evolving data distributions.

Findings

Effective in increasing data volume scenario

Improves performance with growing number of classes

Reduces unnecessary architecture adjustments

Abstract

In real-world applications, data often come in a growing manner, where the data volume and the number of classes may increase dynamically. This will bring a critical challenge for learning: given the increasing data volume or the number of classes, one has to instantaneously adjust the neural model capacity to obtain promising performance. Existing methods either ignore the growing nature of data or seek to independently search an optimal architecture for a given dataset, and thus are incapable of promptly adjusting the architectures for the changed data. To address this, we present a neural architecture adaptation method, namely Adaptation eXpert (AdaXpert), to efficiently adjust previous architectures on the growing data. Specifically, we introduce an architecture adjuster to generate a suitable architecture for each data snapshot, based on the previous architecture and the different extent between current and previous data distributions. Furthermore, we propose an adaptation condition to determine the necessity of adjustment, thereby avoiding unnecessary and time-consuming adjustments. Extensive experiments on two growth scenarios (increasing data volume and number of classes) demonstrate the effectiveness of the proposed method.