Neural Network Inversion For Key Instance Detection In Multiple Instance Learning
- Title
- Neural Network Inversion For Key Instance Detection In Multiple Instance Learning
- Authors
- 신범조
- Date Issued
- 2021
- Publisher
- 포항공과대학교
- Abstract
- Multiple Instance Learning (MIL) involves predicting a single binary label for a bag of instances, given positive or negative labels at bag-level. Since a positive bag contains both positive and negative instances, it is often required to detect positive instances (key instances) when a set of instances is categorized as a positive bag. The attention-based deep MIL model is a recent advance in both bag-level classification and key instance detection (KID). However, if the positive and negative instances in a positive bag are not clearly distinguishable, the attention-based deep MIL model has limited KID performance as the attention scores are skewed to few positive instances. In this paper, we present a method to improve the attention-based deep MIL model in the task of KID. The main idea is to use the neural network inversion to find which instances made contribution to the bag-level prediction produced by the trained MIL model. Moreover, we apply L2 constraint in terms of data into the neural network inversion. Numerical experiments on an MNIST-based image MIL dataset and two real-world histopathology datasets verify the validity of our method, demonstrating the KID performance is significantly improved while the performance of bag-level prediction is maintained.
- URI
- http://postech.dcollection.net/common/orgView/200000366650
https://oasis.postech.ac.kr/handle/2014.oak/111501
- Article Type
- Thesis
- Files in This Item:
- There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.