Pith. sign in

Robust semi-supervised learning via f-divergence and -r \'e nyi divergence

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

1 Pith paper citing it

citation-role summary

background 1

citation-polarity summary

fields

cs.LG 1

years

2025 1

verdicts

CONDITIONAL 1

roles

background 1

polarities

unclear 1

representative citing papers

Loss Functions and Operators Generated by f-Divergences

cs.LG · 2025-01-30 · conditional · novelty 6.0

Using f-divergence regularizers inside Fenchel-Young losses yields convex losses and fast f-softargmax operators, and the α=1.5 variant matches or beats cross-entropy in the tasks tested.

citing papers explorer

Showing 1 of 1 citing paper.

  • Loss Functions and Operators Generated by f-Divergences cs.LG · 2025-01-30 · conditional · none · ref 3

    Using f-divergence regularizers inside Fenchel-Young losses yields convex losses and fast f-softargmax operators, and the α=1.5 variant matches or beats cross-entropy in the tasks tested.