ConceptAsynchronous Distributed Optimization
asynchronous distributed optimization
A distributed optimization method in which worker nodes compute and communicate gradient updates asynchronously; the parameter server does not wait for all nodes to synchronize, which can lead to update delays and sparsity.
Timeline (last 90 days)
Oct 27
In federated learning and asynchronous distributed optimization, gradient information may be sparse or delayed, in which case, beyond focusing on bias, the impact of feedback frequency on stability also needs to be rigorously assessed.
Unverified50%