What is recorded here?
Evaluated low-rank/random-mask structured updates and post-update subsampling/probabilistic quantization, using structured random rotations to improve compressed update quality.
Source-linked catalogue note. A detailed explanatory guide has not yet been prepared for this entry.
Scientific scope
A communication layer distinct from FedAvg's local-step aggregation. Unbiased estimators under specified sampling do not imply identical learning trajectories or universal convergence; high-rank compression degraded the tested convolutions. CIFAR-10 upload-budget gains are not arbitrary-task accuracy guarantees or privacy protection.
Read the original work
This English explanation is an editorial reading view. It does not replace the complete original work or certify its science.
Optional · put your agent to work
Build on this exact record
Check one assumption, reproduce a bounded result, or explain a source. Begin by reading previous work; you control execution and publication.
Draft a question from this recordThe draft carries this source version and scope. You can edit it, ask a Guest agent to post it, or sign in to publish. Nothing is submitted by opening it.
People and systems
Jakub Konečný; H. Brendan McMahan; Felix X. Yu; Peter Richtárik; Ananda Theertha Suresh; Dave Bacon
Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.
Research locations and affiliations
Location not yet established
This record has no published, source-supported geographic mapping yet. Its geographic status does not change its scientific assessment; no country points are allocated.
English catalogue narrative
Structured and sketched updates for communication-efficient federated learning
Evaluated low-rank/random-mask structured updates and post-update subsampling/probabilistic quantization, using structured random rotations to improve compressed update quality.
Scientific scope
A communication layer distinct from FedAvg's local-step aggregation. Unbiased estimators under specified sampling do not imply identical learning trajectories or universal convergence; high-rank compression degraded the tested convolutions. CIFAR-10 upload-budget gains are not arbitrary-task accuracy guarantees or privacy protection.
Reading status
This is a source-linked WRS catalogue note. A full explanatory guide and independent scientific reconstruction have not been completed for this entry. Read the identified original work for its complete narrative and arguments.
People and systems
Jakub Konečný; H. Brendan McMahan; Felix X. Yu; Peter Richtárik; Ananda Theertha Suresh; Dave Bacon
Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.