RC_MF

RC-MF is a two-stage recommender framework that improves biased matrix factorization through regularized residual calibration.

License

The package is licensed under the MIT license.

Installation

Clone the repository and install dependencies:

git clone https://github.com/GAA-UAM/RC_MF.git
cd RC_MF

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

Usage

Basic command

Run experiments from the repository root:

python experiment_runner.py \
    --dataset DATASET_NAME \
    --epochs 50 \
    --seed 0 \
    --models rc_mf

Replace DATASET_NAME with the identifier of a dataset supported by the data loader.

Run RC-MF without Bayesian optimization

When neither --use_bo nor --use_gridsearch is specified, the model is trained using the default parameters defined in:

config/model_search_spaces.yml

Example:

python experiment_runner.py \
    --dataset Beauty \
    --epochs 100 \
    --seed 0 \
    --models rc_mf

Run RC-MF with Bayesian optimization

Use the --use_bo option to tune the model hyperparameters through Bayesian optimization:

python experiment_runner.py \
    --dataset Beauty \
    --epochs 100 \
    --seed 0 \
    --use_bo \
    --models rc_mf

The search space and Bayesian-optimization budget are defined in:

config/model_search_spaces.yml

Documentation

Detailed usage instructions, commands, configuration options, and reproducibility examples are available in the project Wiki.

Citations

If you use RC-MF in your research or work, please consider citing this project using the following citation format.

@article{emami2026residual,
  author  = {Emami, Seyedsaman and Bellog{\'\i}n, Alejandro and Hern{\'a}ndez-Lobato, Daniel},
  title   = {Residual Correction Learning for Matrix Factorization},
  journal = {SSRN Electronic Journal},
  year    = {2026},
  month   = aug,
  note    = {Preprint},
  doi     = {10.2139/ssrn.7273388},
  url     = {https://ssrn.com/abstract=7273388}
}

Authors

• Seyedsaman (Saman) Emami
• Daniel Hernández-Lobato
• Alejandro Bellogín

Grupo de Aprendizaje Automático Universidad Autónoma de Madrid