Meta Input: How to Leverage Off-the-Shelf Deep Neural Networks

TL;DR

This paper introduces a meta input method that enables the effective use of pretrained deep neural networks in new testing environments without modifying the models, by aligning testing data distribution with training data.

Contribution

The paper proposes a novel meta input technique that optimizes additional inputs to adapt pretrained DNNs to different environments without internal model modifications.

Findings

Meta input improves model robustness against environment discrepancies.

The method works with minimal testing data and no internal model changes.

Experimental results demonstrate effectiveness across various tasks.

Abstract

These days, although deep neural networks (DNNs) have achieved a noticeable progress in a wide range of research area, it lacks the adaptability to be employed in the real-world applications because of the environment discrepancy problem. Such a problem originates from the difference between training and testing environments, and it is widely known that it causes serious performance degradation, when a pretrained DNN model is applied to a new testing environment. Therefore, in this paper, we introduce a novel approach that allows end-users to exploit pretrained DNN models in their own testing environment without modifying the models. To this end, we present a \textit{meta input} which is an additional input transforming the distribution of testing data to be aligned with that of training data. The proposed meta input can be optimized with a small number of testing data only by considering the relation between testing input data and its output prediction. Also, it does not require any knowledge of the network's internal architecture and modification of its weight parameters. Then, the obtained meta input is added to testing data in order to shift the distribution of testing data to that of originally used training data. As a result, end-users can exploit well-trained models in their own testing environment which can differ from the training environment. We validate the effectiveness and versatility of the proposed meta input by showing the robustness against the environment discrepancy through the comprehensive experiments with various tasks.