arXiv research

Paper proposes new gradient codes for robust distributed machine learning.

Corrects a mistake in $hv$-block cross-validation consistency claim.

Distributed implementations of gradient-based methods, wherein a server distributes gradient computations across worker machines, suffer from slow running machines, called 'stragglers'. Gradient coding is a coding-theoretic framework to mitigate stragglers by enabling the server to recover the gradient sum in the prese…