README.md

Structured Conformal Inference for Matrix Completion with Applications to Group Recommender Systems

This software repository provides a software implementation of the methods described in the following paper:

"Structured Conformal Inference for Matrix Completion with Applications to Group Recommender Systems"
Ziyi Liang, Tianmin Xie, Xin Tong, Matteo Sesia
arXiv preprint https://arxiv.org/pdf/2404.17561

Paper abstract

We develop a conformal inference method to construct joint confidence regions for structured groups of missing entries within a sparsely observed matrix. This method is useful to provide reliable uncertainty estimation for group-level collaborative filtering; for example, it can be applied to help suggest a movie for a group of friends to watch together. Unlike standard conformal techniques, which make inferences for one individual at a time, our method achieves stronger group-level guarantees by carefully assembling a structured calibration data set mimicking the patterns expected among the test group of interest. We propose a generalized weighted conformalization framework to deal with the lack of exchangeability arising from such structured calibration, and in this process we introduce several innovations to overcome computational challenges. The practicality and effectiveness of our method are demonstrated through extensive numerical experiments and an analysis of the MovieLens 100K data set.

Contents

• smc/ Python package implementing our methods and some alternative benchmarks.
• third_party/ Third-party Python packages imported by our package.
• experiments_real/ Codes to replicate the figures for the experiments with the MovieLens 100K data set.
• experiments_synthetic/ Codes to replicate the figures for the synthetic experiments discussed in the accompanying paper.
• notebooks/ Jupyter notebooks with introductory usage examples.
• dependencies.txt Prerequisites with version number for the smc package.