Replicable algorithms for heterogeneous problems can be composed with O(sum n_i) samples at constant replicability via conversion to perfectly generalizing algorithms, privacy-style composition, and correlated sampling.
arXiv preprint arXiv:2507.11926 , year=
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
New Ω(log n / (log Δ ⋅ polyloglog Δ)) locality lower bound for O(log Δ)-approximate non-signaling dominating set, plus Ω(log n / log Δ) for O(log^β Δ) approximations yielding quantum-LOCAL bounds.
QED bounds cross-run KL divergence in Boltzmann policies by setting temperature proportional to Q-disagreement and reduces return variance by two orders of magnitude on 18 continuous-control tasks without performance loss.
citing papers explorer
-
Replicable Composition
Replicable algorithms for heterogeneous problems can be composed with O(sum n_i) samples at constant replicability via conversion to perfectly generalizing algorithms, privacy-style composition, and correlated sampling.
-
Non-Signaling Locality Lower Bounds for Dominating Set
New Ω(log n / (log Δ ⋅ polyloglog Δ)) locality lower bound for O(log Δ)-approximate non-signaling dominating set, plus Ω(log n / log Δ) for O(log^β Δ) approximations yielding quantum-LOCAL bounds.
-
Behavior-Consistent Deep Reinforcement Learning
QED bounds cross-run KL divergence in Boltzmann policies by setting temperature proportional to Q-disagreement and reduces return variance by two orders of magnitude on 18 continuous-control tasks without performance loss.