Multiple instance learning with bag dissimilarities
作者：
Highlights：
• A general bag dissimilarities framework for multiple instance learning is explored.
• Point set distances and distribution distances are considered.
• Metric dissimilarities are not necessarily more informative.
• Results are competitive with, or outperform state-of-the-art algorithms.
• Practical suggestions for end-users are provided.
摘要
Highlights•A general bag dissimilarities framework for multiple instance learning is explored.•Point set distances and distribution distances are considered.•Metric dissimilarities are not necessarily more informative.•Results are competitive with, or outperform state-of-the-art algorithms.•Practical suggestions for end-users are provided.
论文关键词：Multiple instance learning,Dissimilarity representation,Point set distance,Image classification,Drug activity prediction,Text categorization
论文评审过程：Received 26 June 2013, Revised 26 June 2014, Accepted 18 July 2014, Available online 5 August 2014.
论文官网地址：https://doi.org/10.1016/j.patcog.2014.07.022
