Social, Secure, Scalable, and Efficient
The community building open and collaborative AI anywhere at any scale

An end-to-end machine learning ecosystem for people or organizations to transform their data to intelligence with minimum effort.

Currently, we start from building the AI platform with

  • Cutting-edge federated learning algorithms backed by years of Open Source-oriented research (50+ scientific publications, 1000+ early slack users, and 460+ GitHub forks)
  • Lightweight and cross-platform Edge AI SDK for GPUs, smartphones, and IoTs
  • User-friendly MLOps platform to simplify collaboration and real-world deployment
  • Platform-supported vertical Solutions across a broad range of industries


Wide Adoption by AI Community

FedML open source library has been used widely in the world, including researchers and engineers from the United States, Canada, China, Germany, Denmark, Korea, and Singapore. Some of them are from big companies Google, Amazon, Adobe, Cisco, and Huawei, as well as well-known research-oriented universities such as Stanford, Princeton, USC, HKUST, Tsinghua, etc. They published in top-tier AI conferences including ICML, NeurIPS, ICLR, and AAAI.

3

Products
including open-source, edge AI SDK, and MLOps platform

50+

Scientific Publications
in ML/FL algorithms, security/privacy, systems, and applications

1800+

Open-source Community Users
from all over the world

10+

Industrial Collaborators
from top-tier companies

Our Product and Services

A Versatile Edge-Cloud Ecosystem for Federated Learning/Analytics at Scale
Smoothly transplant from in-lab simulation to real-world system deployment


Open Source

An international community for cutting-edge algorithms


Edge AI SDK

A lightweight and cross-platform design for secure edge training


MLOps Cloud

A user-friendly cloud service for real-world collaboration and deployment



Core Technology is Backed by Years of Cutting-edge Research

Enabling Diverse AI Applications

FedML Ecosystem facilitates federated learning research and productization in diverse application domains. With the foundational support from FedML Core Framework, it supports FedNLP (Natural Language Processing), FedCV (Computer Vision), FedGraphNN (Graph Neural Networks), and FedIoT (Internet of Things).


FedNLP


FedCV


FedGraphNN


FedIoT


Our Team

A Strong Alliance of Industrial and Academia Experience

Salman Avestimehr

Co-Founder and CEO

  • More than 10 years academic, R&D, and management experience in information theory, machine learning and AI, distributed computing, and security/privacy
  • Dean’s Professor of ECE and CS Departments at USC 
  • Director of USC-Amazon Center on Secure & Trusted ML
  • Director of vITAL Research Lab
  • Amazon Scholar in Alexa AI (2020-2022)
  • PhD, UC Berkeley (2008); IEEE Fellow
  • Numerous awards for research, including a PECASE award from the White House (President Obama) and Best Paper Award for FedML.ai 
  • Raised more than $25M in research funding from NSF, DARPA, ONR, ARO, and Industry
  • Homepage: https://www.avestimehr.com/
Chaoyang He

Co-Founder and CTO

  • 10 years R&D experience in top-tier Internet companies:
    Principal Software Engineer and Engineering Manager, Tencent, 2014-2018; Senior Software Engineer, Baidu, 2012-2014; Software Engineer, Huawei, 2011-2012; Part-time researcher with Google/Facebook/Amazon (2018-present)
  • 3 years experience in engineering management and development with business & production-driven strategies for Tencent Cloud, Tencent WeChat Automotive / AI in Car, Tencent Games, Tencent Maps.
  • PhD in CS, University of Southern California
  • Industrial-level Awards: Tencent Outstanding Staff Award, 2015-2016 WeChat MyCar, Special Award for Innovation, 2016; Baidu LBS Group Star Awards, 2013 Q3&Q4
  • Homepage: https://chaoyanghe.com/