Zzong's Notes

❯

❯

collaborative_filtering

❯

Matrix Factorization for Collaborative Filtering Is Just Solving an Adjoint Latent Dirichlet Allocation Model After All

Matrix Factorization for Collaborative Filtering Is Just Solving an Adjoint Latent Dirichlet Allocation Model After All

2026년 8월 29일1 min read

Conclusion

LDA4Rec model, which extends the traditional LDA by incorporating parameters for the popularity of items and conformity of users.

Related

Latent Dirichlet Allocation
categorical distribution
collaborative filtering

References

Paper link

함께 보면 좋은 글

Collaborative Filtering for Implicit Feedback Datasets

Alternating Least Squares Vs.

Probabilistic Matrix Factorization

matrix factorization 은 다음과 같은 확률 모델로 바뀔 수 있다.

Factorization Meets the Item Embedding - Regularizing Matrix Factorization with Item Co-occurrence

Abstract co-factorization model (CoFactor) 제안 item latent factor 를 공유하는 user-item interaction(i.e.

Logistic Matrix Factorization for Implicit Feedback Data

References Paper link .

On the equivalence between Non-negative Matrix Factorization and Probabilistic Latent Semantic Indexing

paper link Abstract & Introduction (Summary) PLSI 와 NMF 는 동일한 목적 함수를 최적화하는 것임을 보임 NMF 의 경우 I-divergence objective function (L 1-normalization NMF) 다만, NMF 와 PLSI 는 서로 다른...

Factorization Machines

combines the advantages of Support Vector Machines (SVM) with factorization models.

Scalable Recommendation with Poisson Factorization

Abstract 본 논문은 hierarchical Poisson matrix factorization(HPF) 모델을 제안. HDF? HDF 는 sparse user & item matrix 를 학습하는데 목적을 두었다.

SimpleX - A Simple and Strong Baseline for Collaborative Filtering

Introduction learning process of collaborative filtering 는 크게 3 개의 요소로 나뉘어진다: 1) interaction encoder, 2)loss functions, 3)negative sampling 적은 negative sampling 과 BPR loss 로...

Variational Autoencoders for Collaborative Filtering

With Mult-VAE, the authors introduce a generative model with multinomial likelihood, propose a different regularization parameter for the learning objective, and use Bayesian...

HybridSVD - When Collaborative Information is Not Enough

Introduction interactions data and side information would be jointly factorized with the help of machine learning/Singular Value Decomposition main contributions generalize...

Conclusion
Related
References