1 Hit in 2.5 sec

InfoBehavior: Self-supervised Representation Learning for Ultra-long Behavior Sequence via Hierarchical Grouping [article]

Runshi Liu, Pengda Qin, Yuhong Li, Weigao Wen, Dong Li, Kefeng Deng, Qiang Wu
2021 arXiv   pre-print
In this paper, we propose a self-supervised method InfoBehavior to automatically extract meaningful representations from ultra-long raw behavior sequences instead of the costly feature selection procedure  ...  Thus, we propose a hierarchical grouping strategy to aggregate ultra-long raw behavior sequences to length-processable high-level embedding sequences.  ...  Therefore, we propose a novel hierarchical grouping strategy to deal with ultra-long sequences.  ... 
arXiv:2106.06905v1 fatcat:qfngohw5vzcp7cawxyhvewp45y