pith. sign in

arXiv preprint arXiv:2502.17323 , year=

When to Forget? Complexity Trade-offs in Machine Unlearning , author= · arXiv 2502.17323

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

1 Pith paper citing it
read on arXiv browse 1 citing papers

fields

cs.LG 1

years

2026 1

verdicts

UNVERDICTED 1

representative citing papers

Exact Unlearning in Reinforcement Learning

cs.LG · 2026-06-02 · unverdicted · novelty 7.0

For any ρ>0 there exists a ρ-TV-stable RL algorithm for tabular MDPs supporting exact unlearning at expected cost ρ√(ln T) of retraining from scratch, with regret O(H²√(SAT)+H³S²A+H^{2.5}S²A/ρ) and matching lower bound Ω(H√(SAT)+SAH/ρ).

citing papers explorer

Showing 1 of 1 citing paper.

  • Exact Unlearning in Reinforcement Learning cs.LG · 2026-06-02 · unverdicted · none · ref 28

    For any ρ>0 there exists a ρ-TV-stable RL algorithm for tabular MDPs supporting exact unlearning at expected cost ρ√(ln T) of retraining from scratch, with regret O(H²√(SAT)+H³S²A+H^{2.5}S²A/ρ) and matching lower bound Ω(H√(SAT)+SAH/ρ).