Provide a concrete introductory example of the PPL in supplementary material.

Pd i=1(1 li) (see (8)), whereli = x>i (X>X) 1xi isthei-thleveragescoreofX. Here, weshow(3).

But despite its wide applicability (see 2.2) and elegance, convergence theory for CCCP is quite limited.

Although various normalization schemes havebeen proposed, theyallfollowacommon theme: normalize acrossspatialdimensions and discard the extracted statistics.

Reviewer#1and#2: Since the original paper submission, we have explored PONO in the context of other applications and model architectures.

To overcome these challenges, we adopt alocal strategy for BO.