Factorization Machines

By Steffen Rendle. 2010

3,071 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Information Systems, Model Architectures, 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

In Information Systems

In Model Architectures

In Recommender Systems

Citations in the corpus

Deep Interest Network for Click-Through Rate Prediction (2017, background): “imension in language modeling.Thismethod,oftenreferredtoasembedding,hasinspired many natural language models and CTR prediction models that need to handle large-scale sparse inputs. LS-PLM [9] and FM [20] models can be viewed as a class of networks with one hidden layer, which first employs embeddin”
VEGA: Towards an End-to-End Configurable AutoML Pipeline (2020, neutral): “Compared with FM (Rendle, 2010) and DeepFM (Guo et al.”
Ask the GRU: Multi-task Learning for Deep Text Recommendations (2016, methodology): “These can be used as features X j in (3) [16, 29].”
Meta-Learning: A Survey (2018, neutral): “based on factorization machines [132] in the first layer, aimed at learning a latent representation for each task to model task similarities.”
VATT: Transformers for Multimodal Self-Supervised Learning from Raw Video, Audio and Text (2021, methodology): “We observe that using matrix factorization on the classifier weight [81] leads to a more stable result across experiments.”
Beyond Bilinear: Generalized Multimodal Factorized High-Order Pooling for Visual Question Answering (2017, methodology): “Inspired by the matrix factorization tricks for uni-modal data [22], [23], the projection matrix Wi in Eq.”
DeepFM: A Factorization-Machine based Neural Network for CTR Prediction (2017, methodology): “Factorization Machines (FM) [Rendle, 2010] model pairwise feature interactions as inner product of latent vectors between features and show very promising results.”
Multi-modal Factorized Bilinear Pooling with Co-attention Learning for Visual Question Answering (2017, methodology): “Inspired by the matrix factorization tricks for uni-modal data [15, 25], the projection matrix W i in Eq.”
StarSpace: Embed All The Things! (2017, methodology): “In the domain of recommendation, embedding models have had a large degree of success, starting from SVD (Goldberg et al. 2001) and its improvements such as SVD++ (Koren and Bell 2015), as well as a host of other techniques, e.g. (Rendle 2010; Lawrence and Urtasun 2009; Shi et al. 2012).”
Personalized Deep Learning for Tag Recommendation (2017, methodology): “Factorization Machine (FM) [16] is an approach that takes advantage of feature engineering and factorization.”
Filter-enhanced MLP is All You Need for Sequential Recommendation (2022, methodology): “The two models largely follow the standard framework of deep sequential models [17, 41, 53], consisting of an embedding layer, a sequence encoder layer and a prediction layer, but adopt RNN and Transformer in the sequence encoder layer, respectively.”
DiffNet++: A Neural Influence and Interest Diffusion Network for Social Recommendation (2020, methodology): “, BPR [36], FM [35]) need to store the embeddings of each user and each item, the space complexity of Q1 is the same as classical embedding based models and grows linearly with users and items.”
Deep Neural Networks and Tabular Data: A Survey (2021, background): “The authors demonstrate that it is possible to replace the handcrafted feature transformations with learned factorization machines (FMs) [100].”
Attentive Collaborative Filtering: Multimedia Recommendation with Item- and Component-Level Attention (2017, methodology): “Since ACF is a generic attention-based CF framework, we plan to extend ACF in various CF models such as factorization machines [29], and the recently proposed Neural CF [20] and Discrete CF [43].”
Heterogeneous Information Network Embedding for Recommendation (2017, methodology): “And Rendle [30] proposes factorization 3 User Movie Director Type Actor Group (a) Douban Movie User Book Author Year Publisher (b) Douban Book User Business Category City Compliment (c) Yelp Fig.”