Optimal augmentation size for BONuS under a subspace prior

From papers

Let m~opt\tilde m_{\textnormal{opt}} be the choice of augmentation size that maximizes the power of BONuS for a subspace prior, with mm fixed. Assume the strong-signal regime

γ2h4>c.\gamma^2h^4>c.

Optimal augmentation-size conjecture. Under this regime,

m~opt=Θ(m).\tilde m_{\textnormal{opt}}=\Theta(\sqrt{m}).

The preceding small-m~\tilde m expansion has the same 1/m~1/\tilde m versus m~/m\tilde m/m trade-off as in the point-mass case, supporting the conjectured scaling. The exact optimizer for BONuS under the subspace prior remains open.

Progress summary

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Sources & referencesView supporting material

Primary source

Abhinav Chakraborty, Junu Lee and Eugene Katsevich, “Power of masking methods for adaptive testing in a multivariate normal means problem”, arXiv:2601.07764 (2026).

Solutions 0

No solutions have been posted yet.