Geodesy

Distributed learning

Quality (philosophy)

World Wide Web

Benchmark (surveying)

Philosophy

Communication

Distillation

Raw data

Federated learning

Pedagogy

Computer science

Distributed computing

Organic chemistry

Psychology

Geography

Models of communication

Epistemology

Sociology

Artificial intelligence

Chemistry

Programming language

Personalization

Machine learning

Nature Communications

Federated learning is a privacy-preserving machine learning technique to train intelligent models from decentralized data, which enables exploiting private data by communicating local model updates in each iteration of model learning rather than the raw data. However, model updates can be extremely large if they contain numerous parameters, and many rounds of communication are needed for model training. The huge communication cost in federated learning leads to heavy overheads on clients and high environmental burdens. Here, we present a federated learning method named FedKD that is both communication-efficient and effective, based on adaptive mutual knowledge distillation and dynamic gradient compression techniques. FedKD is validated on three different scenarios that need privacy protection, showing that it maximally can reduce 94.89% of communication cost and achieve competitive results with centralized model learning. FedKD provides a potential to efficiently deploy privacy-preserving intelligent systems in many scenarios, such as intelligent healthcare and personalization.

Communication-efficient federated learning via knowledge distillation