language-icon Old Web
English
Sign In

COLA: Decentralized Linear Learning

2018 
Decentralized machine learning is a promising emerging paradigm in view of global challenges of data ownership and privacy. We consider learning of linear classification and regression models, in the setting where the training data is decentralized over many user devices, and the learning algorithm must run on-device, on an arbitrary communication network, without a central coordinator. We propose COLA, a new decentralized training algorithm with strong theoretical guarantees and superior practical performance. Our framework overcomes many limitations of existing methods, and achieves communication efficiency, scalability, elasticity as well as resilience to changes in data and participating devices.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    60
    References
    20
    Citations
    NaN
    KQI
    []