DiffVersify: a Scalable Approach to Differentiable Pattern Mining with Coverage Regularization

This paper presents a novel scalable approach to differentiable pattern mining that incorporates coverage regularization. The method enables end-to-end learning of pattern mining tasks while maintaining computational efficiency on large-scale datasets. Our approach demonstrates significant improvements over traditional pattern mining techniques by leveraging modern deep learning optimization strategies.

Conference paper

European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 407-422