Seminar on Privacy Leakage in Federated Learning
10/18/2021
I just gave a seminar on recent advances and open problems on privacy leakage in federated learning.
Self-Supervised FL!
10/12/2021
An essential, but rarely studied, challenge in FL is label deficiency at the edge. This problem is even more pronounced in FL, compared to centralized training, due to the fact that FL users are often reluctant to label their private data and edge devices do not provide an ideal interface to assist with annotation
Rethinking Secure Aggregation in FL!
10/1/2021
Secure model aggregation is a key component of federated learning (FL) that aims at protecting the privacy of each user's individual model while allowing their global aggregation.
JSAIT special issue on distributed learning!
9/30/2021
We have just announced a special issue of the IEEE Journal on Selected Areas in Information Theory (JSAIT) devoted to distributed learning and computing (including federated learning).
Cisco funds FedIoT!
9/27/2021
Cisco research group sponsors our research in Federated Learning for IoT applications.
Congratulations to our Amazon ML Fellows!
8/05/2021
Chaoyang He and Ninareh Mehrabi are the first Amazon ML Fellows in the USC-Amazon Center on Secure and Trusted Machine Learning.
Keynote at the FL-ICML’21
7/27/2021
I just gave a keynote talk on "Securing Secure Aggregation: Mitigating Multi-Round Privacy Leakage in Federated Learning”
Plenary talk at the Conference on Information-Theoretic Cryptography
7/26/2021
I just gave a plenary talk on "Secure Model Aggregation in Federated Learning” at the 2021 Conference on Information-Theoretic Cryptography (ITC).
A Field Guide to Federated Optimization
7/22/2021
In a large collaborative effort, we have just released the paper "A Field Guide to Federated Optimization”
USC-Amazon center projects announced!
7/15/2021
We have just announced five selected research projects to be funded by the USC-Amazon center on Secure and Trusted Machine Learning
Chaoyang He and Saurav Prakash win Qualcomm Innovation Fellowship!
7/1/2021
Congratulations to Chaoyang He and Saurav Prakash for winning a 2021 Qualcomm Innovation Fellowship, for their project on “on-device federated learning”
Welcome To Our Summer Interns For AEOP Program!
6/25/2021
This summer we are hosting four stellar high-school summer interns in the vITAL lab. They will be learning about problems in data science, machine learning, coded computing, and federated learning.
Konica Minolta sponsors FedCV!
6/14/2021
Konica Minolta has just awarded our project for development of FedML for diverse computer vision applications. Looking forward to our collaborations!
SpreadGNN!
6/4/2021
We have just posted a new paper, titled "SpreadGNN: Serverless Multi-task Federated Learning for Graph Neural Networks” with the following abstract:
PipeTransformer at ICML’21
5/8/2021
Our paper "PipeTransformer: Automated Elastic Pipelining for Distributed Training of Transformers" is accepted for presentation at ICML'21.
FL Seminar
4/25/2021
Here is a video of my recent talk highlighting several recent results from our group on various aspects of Federate Learning.