# Blockwise Feature Interaction in Recommendation Systems

arXiv: https://arxiv.org/abs/2306.15881 · HTML: https://mlsanity.com/paper/2306.15881/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 19 references that are in the map.
- First posted on arXiv 28 Jun 2023.

## Builds on (most cited first, 9 of 9)

- [DeepFM: A Factorization-Machine based Neural Network for CTR Prediction](https://mlsanity.com/paper/1703.04247/) (2017)
- [Text and Code Embeddings by Contrastive Pre-Training](https://mlsanity.com/paper/2201.10005/) (2022)
- [DCN V2: Improved Deep & Cross Network and Practical Lessons for Web-scale Learning to Rank Systems](https://mlsanity.com/paper/2008.13535/) (2020)
- [xDeepFM: Combining Explicit and Implicit Feature Interactions for Recommender Systems](https://mlsanity.com/paper/1803.05170/) (2018)
- [DeCLUTR: Deep Contrastive Learning for Unsupervised Textual Representations](https://mlsanity.com/paper/2006.03659/) (2020)
- [Pre-training Tasks for Embedding-based Large-scale Retrieval](https://mlsanity.com/paper/2002.03932/) (2020)
- [General Multi-label Image Classification with Transformers](https://mlsanity.com/paper/2011.14027/) (2020)
- [Deep Fragment Embeddings for Bidirectional Image Sentence Mapping](https://mlsanity.com/paper/1406.5679/) (2014)
- [Hierarchical Bilinear Pooling for Fine-Grained Visual Recognition](https://mlsanity.com/paper/1807.09915/) (2018)

## Abstract (arXiv)

Feature interactions can play a crucial role in recommendation systems as they capture complex relationships between user preferences and item characteristics. Existing methods such as Deep & Cross Network (DCNv2) may suffer from high computational requirements due to their cross-layer operations. In this paper, we propose a novel approach called blockwise feature interaction (BFI) to help alleviate this issue. By partitioning the feature interaction process into smaller blocks, we can significantly reduce both the memory footprint and the computational burden. Four variants (denoted by P, Q, T, S, respectively) of BFI have been developed and empirically compared. Our experimental results demonstrate that the proposed algorithms achieves close accuracy compared to the standard DCNv2, while greatly reducing the computational overhead and the number of parameters. This paper contributes to the development of efficient recommendation systems by providing a practical solution for improving feature interaction efficiency.