CorrFL: Correlation-based Neural Network Architecture for Unavailability Concerns in a Heterogeneous IoT Environment.

This code provides the implementation of a Correlation Neural Network in Federated Learning to address model heterogeneity and unavailability concern in the Federated Learning environment. All the code documentation and variable definition is in accordance with the content of the manuscript published in IEEE Transactions on Network and Service Management: I. Shaer and A. Shami, "CorrFL: Correlation-based Neural Network Architecture for Unavailability Concerns in a Heterogeneous IoT Environment, " IEEE Transactions on Network and Service Management, doi: 10.1109/TNSM.2023.3278937.

Before experimenting with the code, the following steps need to be implemented:

1. Download the data employed for this work, which can be found using this link: https://zenodo.org/record/3774723#.ZGEhAHbMKUl. Since we only ran out experiments on room00, retain the data of this room, which can be found by following data_cleaning/dirty_room/room00 directory.
2. Create the datasets folder in the root directory (on the same level as src) and its sub-folders dirty_data
3. Relocate the remaining files in the directory datasets/dirty_data/room00_preprocessed, which is defined in the dataset_dir variable in src/generate_granular_data.py. Make sure to rename the csv files to their corresponding node names. For example, room00_THP-CO2_924_20190101-20191231.csv to node_924.
4. Create datasets/dirty_data/room00_gran5_fine folder directory as declared in generate_granular_data.py.
5. Run the generate_granular_data.py using the following command: python generate_granular_data.py. This results in creating datasets in the datasets/dirty_data/room00_gran5_fine folder, which will be used throughout the experiments.
6. Create the results folder in the root directory and its sub-folders figures, insights, models, and stats
7. Run driver.py using the following command python driver.py.

This paper highlights the problem of Oblique Federated Learning, which is halfway the Vertical Federated Learning and Horizontal Federated Learning, featuring non-uniformity in the feature space of local agents while these local agents share some and not all of the feature space. Therefore, this paper and code expose a novel problem in the space of Federated Learning and propose exciting research questions for practitioners and researchers to address.

Requirements

The requirements are included in the requirements.txt file. To install the packages included in this file, use the following command: pip install -r requirements.txt

Contact-Info

Please feel free to contact me for any questions or research opportunities.

• Email: [email protected]
• Gihub: https://github.com/ibrahimshaer and https://github.com/Western-OC2-Lab
• LinkedIn: Ibrahim Shaer
• Google Scholar: Ibrahim Shaer and OC2 Lab