FIVES: Feature Interaction Via Edge Search for Large-Scale Tabular Data

  title={FIVES: Feature Interaction Via Edge Search for Large-Scale Tabular Data},
  author={Yuexiang Xie and Zhen Wang and Yaliang Li and Bolin Ding and Nezihe Merve Gurel and Ce Zhang and Minlie Huang and Wei Lin and Jingren Zhou},
  journal={Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery \& Data Mining},
  • Yuexiang Xie, Zhen Wang, +6 authors Jingren Zhou
  • Published 2021
  • Computer Science, Mathematics
  • Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining
High-order interactive features capture the correlation between different columns and thus are promising to enhance various learning tasks on ubiquitous tabular data. To automate the generation of interactive features, existing works either explicitly traverse the feature space or implicitly express the interactions via intermediate activations of some designed models. These two kinds of methods show that there is essentially a trade-off between feature interpretability and search efficiency… Expand

Figures and Tables from this paper

AutoML: A Perspective where Industry Meets Academy
Machine learning methods have been adopted for various real-world applications, ranging from social networks, online image/video-sharing platforms, and e-commerce to education, healthcare, etc.… Expand


xDeepFM: Combining Explicit and Implicit Feature Interactions for Recommender Systems
A novel Compressed Interaction Network (CIN), which aims to generate feature interactions in an explicit fashion and at the vector-wise level and is named eXtreme Deep Factorization Machine (xDeepFM), which is able to learn certain bounded-degree feature interactions explicitly and can learn arbitrary low- and high-order feature interactions implicitly. Expand
Deep & Cross Network for Ad Click Predictions
This paper proposes the Deep & Cross Network (DCN), which keeps the benefits of a DNN model, and beyond that, it introduces a novel cross network that is more efficient in learning certain bounded-degree feature interactions. Expand
AutoInt: Automatic Feature Interaction Learning via Self-Attentive Neural Networks
An effective and efficient method called the AutoInt to automatically learn the high-order feature interactions of input features and map both the numerical and categorical features into the same low-dimensional space is proposed. Expand
Towards Automated Neural Interaction Discovery for Click-Through Rate Prediction
This work proposes an automated interaction architecture discovering framework for CTR prediction named AutoCTR, which performs evolutionary architecture exploration with learning-to-rank guidance at the architecture level and achieves acceleration using low-fidelity model. Expand
Feature Interaction Interpretability: A Case for Explaining Ad-Recommendation Systems via Neural Interaction Detection
This work proposes a method to both interpret and augment the predictions of black-box recommender systems, in particular, to interpret feature interactions from a source recommender model and explicitly encode these interactions in a targetRecommender model, where both source and target models are black-boxes. Expand
AutoLearn β€” Automated Feature Generation and Selection
This work presents AutoLearn, a regression-based feature learning algorithm, that requires no domain knowledge and is hence generic, and can improve the overall prediction accuracy by 13.28%, compared to original feature space and 5.87% over other top performing models. Expand
DCN-M: Improved Deep & Cross Network for Feature Cross Learning in Web-scale Learning to Rank Systems
The improved DCN-M is more expressive yet remains cost efficient at feature interaction learning, especially when coupled with a mixture of low-rank architecture, and outperforms all the state-of-the-art algorithms on popular benchmark datasets. Expand
DeepFM: A Factorization-Machine based Neural Network for CTR Prediction
This paper shows that it is possible to derive an end-to-end learning model that emphasizes both low- and high-order feature interactions, and combines the power of factorization machines for recommendation and deep learning for feature learning in a new neural network architecture. Expand
Wide & Deep Learning for Recommender Systems
Wide & Deep learning is presented---jointly trained wide linear models and deep neural networks---to combine the benefits of memorization and generalization for recommender systems and is open-sourced in TensorFlow. Expand
Deep Crossing: Web-Scale Modeling without Manually Crafted Combinatorial Features
The Deep Crossing model is proposed which is a deep neural network that automatically combines features to produce superior models and was able to build, from scratch, two web-scale models for a major paid search engine, and achieve superior results with only a sub-set of the features used in the production models. Expand