Learning Maximally Predictive Prototypes in Multiple Instance Learning
arXiv:1910.00965
Abstract
In this work, we propose a simple model that provides permutation invariant maximally predictive prototype generator from a given dataset, which leads to interpretability of the solution and concrete insights to the nature and the solution of a problem. Our aim is to find out prototypes in the feature space to map the collection of instances (i.e. bags) to a distance feature space and simultaneously learn a linear classifier for multiple instance learning (MIL). Our experiments on classical MIL benchmark datasets demonstrate that proposed framework is an accurate and efficient classifier compared to the existing approaches.
arXiv commentSets & Partitions Workshop at NeurIPS 2019
References in corpus (8)
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- Attention-based Deep Multiple Instance Learning
- Revisiting Multiple Instance Neural Networks
- Layer Normalization
- Multiple Instance Learning with Bag Dissimilarities
- Dissimilarity-based Ensembles for Multiple Instance Learning
- PCL: Proposal Cluster Learning for Weakly Supervised Object Detection
- Adaptive pooling operators for weakly labeled sound event detection