TY - JOUR
T1 - SALE
T2 - Self-adaptive LSH encoding for multi-instance learning
AU - Xu, Dongkuan
AU - Wu, Jia
AU - Li, Dewei
AU - Tian, Yingjie
AU - Zhu, Xingquan
AU - Wu, Xindong
PY - 2017/11/1
Y1 - 2017/11/1
N2 - Multi-instance learning (MIL) is commonly used to classify a set of instances, also known as a bag, where labels for the training set are only available for each bag. Many MIL methods exist, but they often suffer from high computation complexity and the key information from MIL being ignored, which deteriorates the classification performance. Recently, locality-sensitive hashing (LSH), with its high scalability, has shown the ability in enhancing MIL performance. However, for these LSH-based methods, the fixed number of bits is used to represent each projected dimension, resulting in subtle information loss and the algorithm performance reduction. In this paper, we propose a self-adaptive LSH encoding method for MIL, termed as SALE. SALE uses LSH to generate the primary batches, followed by a self-adaptive process for reconstruction. Reconstructed bags are transformed into random super histograms (RSH) using an incomplete coding method, and then weighted through a scheme that takes advantage of key instances. These weighted RSHs are used to train the learning model. SALE efficiently deals with large MIL problems, due to its low complexity and RSH's ability to exploit key information of MIL. Experiments demonstrate SALE's good performance compared to state-of-the-art MIL methods.
AB - Multi-instance learning (MIL) is commonly used to classify a set of instances, also known as a bag, where labels for the training set are only available for each bag. Many MIL methods exist, but they often suffer from high computation complexity and the key information from MIL being ignored, which deteriorates the classification performance. Recently, locality-sensitive hashing (LSH), with its high scalability, has shown the ability in enhancing MIL performance. However, for these LSH-based methods, the fixed number of bits is used to represent each projected dimension, resulting in subtle information loss and the algorithm performance reduction. In this paper, we propose a self-adaptive LSH encoding method for MIL, termed as SALE. SALE uses LSH to generate the primary batches, followed by a self-adaptive process for reconstruction. Reconstructed bags are transformed into random super histograms (RSH) using an incomplete coding method, and then weighted through a scheme that takes advantage of key instances. These weighted RSHs are used to train the learning model. SALE efficiently deals with large MIL problems, due to its low complexity and RSH's ability to exploit key information of MIL. Experiments demonstrate SALE's good performance compared to state-of-the-art MIL methods.
KW - Locality-sensitive hashing
KW - Machine learning
KW - Multi-instance learning
KW - Self-adaptive learning
UR - http://www.scopus.com/inward/record.url?scp=85019444056&partnerID=8YFLogxK
U2 - 10.1016/j.patcog.2017.04.029
DO - 10.1016/j.patcog.2017.04.029
M3 - Article
AN - SCOPUS:85019444056
SN - 0031-3203
VL - 71
SP - 460
EP - 482
JO - Pattern Recognition
JF - Pattern Recognition
ER -