// MODEL OPTIMIZATION AND PROMPT SYNTAX TERM
Federated Learning
A way to train AI models on many devices without centralizing the raw data, keeping it private.
TECHNICAL DEFINITION
A distributed machine learning paradigm where models are trained collaboratively across multiple decentralized edge devices or servers holding local data samples, without exchanging the raw data itself, thus enhancing data privacy.
BACKGROUND
Adversarial machine learning is the study of the attacks on machine learning algorithms, and of the defenses against such attacks.
READ MORE ON WIKIPEDIASYNONYMS & ALIASES
- Collaborative learning
- distributed ML
- privacy-preserving ML
USAGE NOTE
Federated learning allows mobile devices to contribute to model training while keeping user data on-device.
DEVELOPERS
Organizations developing technology related to Federated Learning.
Pioneered federated learning and actively develops frameworks like TensorFlow Federated, applying it to various products like Gboard for on-device model training while preserving user privacy.
Develops the NVIDIA FLARE (Federated Learning Application Runtime Environment) platform and SDKs, enabling secure, privacy-preserving AI collaboration across distributed data sources, especially in healthcare and other industries.
Conducts extensive research and offers enterprise solutions for privacy-preserving AI, including federated learning, to allow organizations to train models collaboratively without sharing raw data.
Engages in federated learning research and integrates privacy-preserving machine learning techniques into its Azure AI services, enabling distributed model training for various enterprise applications.
An open-source community dedicated to building tools for secure, privacy-preserving AI, with a strong focus on federated learning and other privacy-enhancing technologies through libraries like PySyft.
Develops hardware-optimized solutions and software toolkits that support privacy-preserving AI, including federated learning, often leveraging secure enclaves and other technologies to protect data during collaborative training.
Utilizes on-device machine learning with privacy-preserving techniques, often employing federated learning-like principles for improving models based on user data without compromising individual privacy.
Actively researches and applies federated learning across its various business units, particularly in finance, healthcare, and e-commerce, to enable secure and collaborative AI model development.