DeepFM: A Factorization-Machine based Neural Network for CTR Prediction
By Huifeng Guo, Ruiming Tang, Yunming Ye, Zhenguo Li, Xiuqiang He. 2017
3,129 citations recorded by the source corpus.
Learning sophisticated feature interactions behind user behaviors is critical in maximizing CTR for recommender systems. Despite great progress, existing methods seem to have a strong bias towards low- or high-order interactions, or require expertise feature engineering. In this paper, we show that it is possible to derive an end-to-end learning model that emphasizes both low- and high-order feature interactions. The proposed model, DeepFM, combines the power of factorization machines for recommendation and deep learning for feature learning in a new neural network architecture. Compared to the latest Wide & Deep model from Google, DeepFM has a shared input to its "wide" and "deep" parts, with no need of feature engineering besides raw features. Comprehensive experiments are conducted to demonstrate the effectiveness and efficiency of DeepFM over the existing models for CTR prediction, on both benchmark data and commercial data.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Information Systems, Recommender Systems
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
In Computer Vision and Pattern Recognition
- Dacheng Tao
- Luc Van Gool
- Xiaogang Wang
- Yi Yang
- Alan L. Yuille
- Trevor Darrell
- Bernt Schiele
- Ming-Hsuan Yang
In Information Systems
In Recommender Systems
Citations in the corpus
- DKN: Deep Knowledge-Aware Network for News Recommendation (2018, methodology): “• DeepFM [13] is also a general deep model for recommendation, which combines a component of factorization machines and a component of deep neural networks that share the input.”
- Deep Interest Network for Click-Through Rate Prediction (2017, methodology): “tation vector for the instance. MLP. Given the concatenated dense representation vector, fully connected layers are used to learn the combination of features automatically. Recently developed methods [4, 5, 10] focus on designing structures of MLP for better information extraction. Loss. The objecti”
- VEGA: Towards an End-to-End Configurable AutoML Pipeline (2020, background): “…intelligence in various tasks, such as computer vision (He et al., 2016; Ren et al., 2015), natural language processing (NLP) (Yin et al., 2017; Devlin et al., 2018), autonomous driving (Al-Qizwini et al., 2017; Kuutti et al., 2020) and recommender systems (Guo et al., 2017; Liu et al., 2020a).”
- AutoEmb: Automated Embedding Dimensionality Search in Streaming Recommendations (2020, methodology): “Guo et al.[8] proposed an end-toend model named DeepFM to integrate factorization machine and Multilayer Perceptron (MLP) seamlessly.”
- Merlin: A GPU Accelerated Recommendation Framework (2020, methodology): “models including Wide and Deep[3], Deep Cross Networks[14], DeepFM[6], and DLRM[8] to enable fast experimentation and pro-”
- Understanding Capacity-Driven Scale-Out Neural Recommendation Inference (2020, background): “However, more recent recommender systems have used deep neural networks to combine a variety of dense and sparse input features into a more generalized predictive model [1], [3], [22]–[25].”
- Deep Distributed Fusion Network for Air Quality Prediction (2018, methodology): “ DeepFM [17]: Factorization-machine based neural network, modeling both high-order feature interactions and low-order feature interactions.”
- A Neural Influence Diffusion Model for Social Recommendation (2019, methodology): “interactions between user and item embeddings [12]. As sometimes the user and item features are sparse, many deep learning based models have been proposed to tackle how to model these sparse features [9, 38]. In contrast to these works, we do not consider the scenario of sparse features and put emph”
- Deep Learning Based Recommender System (2017, methodology): “DeepFM [47] is an end-to-end model which seamlessly integrates factorization machine and MLP.”
- Recommendation as Language Processing (RLP): A Unified Pretrain, Personalized Prompt & Predict Paradigm (P5) (2022, background): “Recently, deep neural network models [3, 4, 18, 70] facilitate crossing and combination among even more diverse and sophisticated features.”
- xDeepFM: Combining Explicit and Implicit Feature Interactions for Recommender Systems (2018, methodology): “Representative models include FNN [44], PNN [30], DeepCross [36], NFM [11], DCN [39], Wide&Deep [4], and DeepFM [8].”
- Why do tree-based models still outperform deep learning on tabular data? (2022, neutral): “Deep learning for tabular data As described by Borisov et al. [2021] in their review of the field, there have been various attempts to make deep learning work on tabular data: data encoding techniques to make tabular data better suited for deep learning [Hancock and Khoshgoftaar, 2020, Yoon et al., 2”
- CPR: Understanding and Improving Failure Tolerant Training for Deep Learning Recommendation with Partial Recovery (2020, methodology): “…recommendation algorithms, from classical techniques (Van Meteren & Van Someren, 2000; Sarwar et al., 2001; Koren et al., 2009) to machine learning (He et al., 2017; Wang et al., 2016; Rendle & Schmidt-Thieme, 2010) and deep learning (Cheng et al., 2016; Naumov et al., 2019; Zhou et al., 2018;…”
- Estimation-Action-Reflection: Towards Deep Interaction Between Conversational and Recommender Systems (2020, background): “Neural FM [16] and DeepFM [15] have improved FM’s representation ability with deep neural networks.”
- Big Data Service Architecture: A Survey (2020, background): “With the in-depth development of big data technologies, the machine learning is widely used for the in-depth big data analysis [54-56].”
- Progressive Layered Extraction (PLE): A Novel Multi-Task Learning (MTL) Model for Personalized Recommendations (2020, background): “1% increase of AUC or MSE contributes significant improvement to online metrics in our system, which is also mentioned in [4, 6, 14].”
- Multi-Interest Network with Dynamic Routing for Recommendation at Tmall (2019, background): “Neural Collaborative Filtering (NCF) [11], DeepFM [9] and Deep Matrix Factorization Models (DMF) [27] construct a neural network composed of several MLPs to model the interaction between users and items.”