Federated learning

<p>Federated learning (also known as collaborative learning) is a <a href="/facts/Machine_learning/e0w0XJTu">machine learning</a> technique in a setting where multiple entities (often called clients) collaboratively train a model while keeping their data <a href="/facts/Decentralized_computing/oTp3d3XL">decentralized</a>, rather than centrally stored. A defining characteristic of federated learning is data <a href="/facts/Homogeneity_and_heterogeneity_(statistics)/gok3LHv3">heterogeneity</a>. Because client data is decentralized, data samples held by each client may not be <a href="/facts/Independent_and_identically_distributed_random_variables/othIRaWt">independently and identically distributed</a>.
</p><p>Federated learning is generally concerned with and motivated by issues such as <a href="/facts/Information_Privacy/fb6DMpdN">data privacy</a>, <a href="/facts/Data_minimization/AwKRdbcK">data minimization</a>, and data access rights. Its applications involve a variety of research areas including <a href="/facts/Arms_industry/A8mywI9w">defence</a>, <a href="/facts/Telecommunications/vxoj1cxR">telecommunications</a>, the <a href="/facts/Internet_of_things/rwwGAxn3">Internet of things</a>, and <a href="/facts/Pharmaceutical_industry/5mCITYys">pharmaceuticals</a>.
</p>

Federated learning open-in-new

Federated learning