Nested multi-instance classification

TL;DR

This paper introduces a nested multi-instance deep network for classifying groups of data instances, such as images, with novel methods for handling missing instances and improving training robustness, demonstrating superior performance on real-world datasets.

Contribution

It presents a generic nested multi-instance network architecture with innovative missing data handling and manual dropout techniques, enhancing classification accuracy over existing methods.

Findings

Outperforms baseline methods on real-world datasets

Effective handling of missing instances improves accuracy

Manual dropout increases training robustness

Abstract

There are classification tasks that take as inputs groups of images rather than single images. In order to address such situations, we introduce a nested multi-instance deep network. The approach is generic in that it is applicable to general data instances, not just images. The network has several convolutional neural networks grouped together at different stages. This primarily differs from other previous works in that we organize instances into relevant groups that are treated differently. We also introduce a method to replace instances that are missing which successfully creates neutral input instances and consistently outperforms standard fill-in methods in real world use cases. In addition, we propose a method for manual dropout when a whole group of instances is missing that allows us to use richer training data and obtain higher accuracy at the end of training. With specific pretraining, we find that the model works to great effect on our real world and pub-lic datasets in comparison to baseline methods, justifying the different treatment among groups of instances.