Day 1
Last summer, I changed my learning journey from academic research to developer. However, I always wanted to continue my research on #FederatedLearning. But given the job schedule and traveling, it is difficult to have time for research.
On my social feed, i came to know about #30DaysOfFLCode (https://info.openmined.org/30daysofflcode) from #OpenMined, which encourage to devote 1 hour daily on learning #federatedLearning and #privacyEnhancingTechnologies.
With this background, I am committing myself for learning and organizing content and resources for #federatedlearning research for next 30 days and beyond.
As preparation, I have created a github repo (https://github.com/urwithajit9/30DaysOfFLCode) that will have all resources related to #federatedlearning that i will use or collect during this new journey and new way of focused learning.
Day 1 : Organizing resources and setting the environment
I have read many publications on #federatedlearning in the past and here are my three starting papers and always go to papers:
Shokri, R., & Shmatikov, V. (2015, October). Privacy-preserving deep learning. In Proceedings of the 22nd ACM SIGSAC conference on computer and communications security (pp. 1310-1321). Link
McMahan, Brendan, et al. "Communication-efficient learning of deep networks from decentralized data." Artificial intelligence and statistics. PMLR, 2017. Link
Kairouz, Peter, et al. "Advances and open problems in federated learning." Foundations and trends® in machine learning 14.1–2 (2021): 1-210. Link
Platforms:
I will be exploring three FL platforms mainly, along with other libraries:
Scaleout Systems Link
PySyft: An open-source library developed by OpenMined that provides tools for building secure, privacy-preserving federated learning systems using PyTorch. Link
Flower: An open-source framework developed by Adap, which allows you to build federated learning systems using a variety of machine learning libraries.Link
I am happy and thoughtful for the journey ahead and to participate in the #30DaysOfFLCode challenge. Let’s learn and improve #deeplearning with #privacy !!!
Top comments (0)