Pith. sign in

On the nystrom approximation for preconditioning in kernel machines

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

stat.ML 1

years

2024 1

verdicts

CONDITIONAL 1

representative citing papers

Fast training of large kernel models with delayed projections

stat.ML · 2024-11-25 · conditional · novelty 6.0

Delaying projections in EigenPro-style kernel training cuts the amortized per-batch cost from quadratic to linear in model size, yielding up to hundreds of times faster training at comparable accuracy.

citing papers explorer

Showing 1 of 1 citing paper.

  • Fast training of large kernel models with delayed projections stat.ML · 2024-11-25 · conditional · none · ref 3

    Delaying projections in EigenPro-style kernel training cuts the amortized per-batch cost from quadratic to linear in model size, yielding up to hundreds of times faster training at comparable accuracy.