Gaussian Process Assisted Meta-learning for Image Classification and Object Detection Models

TL;DR

This paper introduces a Gaussian process-based meta-learning framework that guides data collection for image classification and object detection, improving model performance by strategically selecting training data based on metadata.

Contribution

It proposes a novel GP-assisted meta-learning method to optimize data acquisition for machine learning models using metadata, enhancing performance over random data collection.

Findings

GP surrogate effectively models response surface for data acquisition

Meta-learning approach outperforms random data selection

Application to aerial imagery improves object detection accuracy

Abstract

Collecting operationally realistic data to inform machine learning models can be costly. Before collecting new data, it is helpful to understand where a model is deficient. For example, object detectors trained on images of rare objects may not be good at identification in poorly represented conditions. We offer a way of informing subsequent data acquisition to maximize model performance by leveraging the toolkit of computer experiments and metadata describing the circumstances under which the training data was collected (e.g., season, time of day, location). We do this by evaluating the learner as the training data is varied according to its metadata. A Gaussian process (GP) surrogate fit to that response surface can inform new data acquisitions. This meta-learning approach offers improvements to learner performance as compared to data with randomly selected metadata, which we illustrate on both classic learning examples, and on a motivating application involving the collection of aerial images in search of airplanes.