A two-stage policy-aware factorial experiment design models outcomes as low-rank tensors, applies tensor completion and sequential halving to identify high-performing policies, and provides regret bounds that scale with tensor degrees of freedom rather than full combinatorial size.
large ℓ∞ error
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Policy-Aware Design of Large-Scale Factorial Experiments
A two-stage policy-aware factorial experiment design models outcomes as low-rank tensors, applies tensor completion and sequential halving to identify high-performing policies, and provides regret bounds that scale with tensor degrees of freedom rather than full combinatorial size.