Layered convolutional dictionary learning for sparse coding itemsets

摘要: Dictionary learning for sparse coding has been successfully used in different domains, however, never employed the interesting itemset mining. In this paper, we formulate an optimization problem extracting a representation of itemsets and show that discrete nature makes it NP-hard. An efficient approximation algorithm is presented which greedily solves maximum set cover to reduce overall compression loss. Furthermore, incorporate our into layered convolutional model learn nonredundant dictionary items. Following intuition deep learning, approach convolves learned items discovers statistically dependent patterns using chi-square hierarchical fashion; each layer having more abstract compressed than previous. extensive empirical validation performed on thirteen datasets, showing better interpretability semantic coherence two existing state-of-the-art methods.