Reinforcement learning search makes messy spreadsheet data easier to predict

Original: Hierarchical and Permutation-Invariant Feature Transformation Learning via Policy-Guided Embedding Search

Source: arxiv.org ↗