# GLocal-K: Global and Local Kernels for Recommender Systems

arXiv: https://arxiv.org/abs/2108.12184 · HTML: https://mlsanity.com/paper/2108.12184/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 13 references that are in the map.
- First posted on arXiv 27 Aug 2021.

## Builds on (most cited first, 4 of 4)

- [Geometric Matrix Completion with Recurrent Multi-Graph Neural Networks](https://mlsanity.com/paper/1704.06803/) (2017)
- [Inductive Matrix Completion Based on Graph Neural Networks](https://mlsanity.com/paper/1904.12058/) (2019)
- [A Neural Autoregressive Approach to Collaborative Filtering](https://mlsanity.com/paper/1605.09477/) (2016)
- [Scalable Probabilistic Matrix Factorization with Graph-Based Priors](https://mlsanity.com/paper/1908.09393/) (2019)

## Built on by (most cited first, 11 of 11)

- [Neural Processes with Stochastic Attention: Paying more attention to the context dataset](https://mlsanity.com/paper/2204.05449/) (2022)
- [Cluster-Enhanced Federated Graph Neural Network for Recommendation](https://mlsanity.com/paper/2412.08066/) (2024)
- [RSAttAE: An Information-Aware Attention-based Autoencoder Recommender System](https://mlsanity.com/paper/2502.06705/) (2025)
- [SUPER-Rec: SUrrounding Position-Enhanced Representation for Recommendation](https://mlsanity.com/paper/2209.04154/) (2022)
- [Multi-Metric AutoRec for High Dimensional and Sparse User Behavior Data Prediction](https://mlsanity.com/paper/2212.13879/) (2022)
- [Ae2I: A Double Autoencoder for Imputation of Missing Values](https://mlsanity.com/paper/2301.06633/) (2023)
- [Unveiling Optimal SDG Pathways: An Innovative Approach Leveraging Graph Pruning and Intent Graph for Effective Recommendations](https://mlsanity.com/paper/2309.11741/) (2023)
- [Beyond RMSE and MAE: Introducing EAUC to unmask hidden bias and unfairness in dyadic regression models](https://mlsanity.com/paper/2401.10690/) (2024)
- [What Are We Optimizing For? A Human-centric Evaluation of Deep Learning-based Movie Recommenders](https://mlsanity.com/paper/2401.11632/) (2024)
- [Federated Learning based Latent Factorization of Tensors for Privacy-Preserving QoS Prediction](https://mlsanity.com/paper/2407.19828/) (2024)
- [BPL: Bias-adaptive Preference Distillation Learning for Recommender System](https://mlsanity.com/paper/2510.16076/) (2025)

## Abstract (arXiv)

Recommender systems typically operate on high-dimensional sparse user-item matrices. Matrix completion is a very challenging task to predict one's interest based on millions of other users having each seen a small subset of thousands of items. We propose a Global-Local Kernel-based matrix completion framework, named GLocal-K, that aims to generalise and represent a high-dimensional sparse user-item matrix entry into a low dimensional space with a small number of important features. Our GLocal-K can be divided into two major stages. First, we pre-train an auto encoder with the local kernelised weight matrix, which transforms the data from one space into the feature space by using a 2d-RBF kernel. Then, the pre-trained auto encoder is fine-tuned with the rating matrix, produced by a convolution-based global kernel, which captures the characteristics of each item. We apply our GLocal-K model under the extreme low-resource setting, which includes only a user-item rating matrix, with no side information. Our model outperforms the state-of-the-art baselines on three collaborative filtering benchmarks: ML-100K, ML-1M, and Douban.