Higher-Order Factorization Machines
This work addresses a bottleneck in machine learning for high-dimensional data with feature interactions, offering incremental improvements in efficiency and model optimization.
The paper tackles the lack of efficient training algorithms for higher-order factorization machines (HOFMs) by presenting the first generic and efficient algorithms for arbitrary-order HOFMs, along with variants that reduce model size and prediction times while maintaining similar accuracy, as demonstrated on four link prediction tasks.
Factorization machines (FMs) are a supervised learning approach that can use second-order feature combinations even when the data is very high-dimensional. Unfortunately, despite increasing interest in FMs, there exists to date no efficient training algorithm for higher-order FMs (HOFMs). In this paper, we present the first generic yet efficient algorithms for training arbitrary-order HOFMs. We also present new variants of HOFMs with shared parameters, which greatly reduce model size and prediction times while maintaining similar accuracy. We demonstrate the proposed approaches on four different link prediction tasks.