Latent variable model for high-dimensional point process with structured missingness

This repository contains the python scripts for our paper published in the Proceedings of The Forty-First International Conference on Machine Learning (ICML).

Repository structure

The repository contains two main folders - LLSM and LLPPSM which contain implementation and experimental configs to the corresponding variants of our method. More information regarding these two method variants is in the manuscript.

Poster

Prerequisites

• Python
• PyTorch
• GPyTorch
• Torchvision
• Pandas
• Matplotlib

Downloading MNIST digits

• Download and unzip archive from here: https://www.dropbox.com/s/j80vfwcqqu3vmnf/trainingSet.tar?dl=0

Generating Health MNIST experiment data

• To create training/test data, labels as well as mask for LLSM, go to the corresponding folder and run: python Health_MNIST_generate.py --source=./trainingSet --destination=./data --num_3=10 --num_6=10 --missing=25 --data_file_name=data.csv --labels_file_name=labels.csv --mask_file_name=mask.csv --data_masked_file_name=masked_data.csv
• To create training/test data, labels as well as mask for LLPPSM, go to the corresponding folder and run: Health_MNIST_generate.py --source=./trainingSet --destination=./data --num_3=10 --num_6=10 --missing=25 --data_file_name=data.csv --labels_file_name=labels.csv --mask_file_name=mask.csv --data_masked_file_name=masked_data.csv --D=15
• See Health_MNIST_generate.py for configuration in both cases

Training

• To run training for LLSM, go to the corresponding folder and run: python LVAE.py --f=./config/LLSM.txt
• To run training for LLPPSM, go to the corresponding folder and run: python LVAE.py --f=./config/LLPPSM.txt
• If you want to use a custom dataset, generate data and labels with accordance to .csv files in Health MNIST and modify config files

Cite

Please cite this work as:

@InProceedings{pmlr-v235-sinelnikov24a, title = {Latent variable model for high-dimensional point process with structured missingness}, author = {Sinelnikov, Maksim and Haussmann, Manuel and L"{a}hdesm"{a}ki, Harri}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {45525--45543}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/sinelnikov24a/sinelnikov24a.pdf}, url = {https://proceedings.mlr.press/v235/sinelnikov24a.html} }