Similarly the ISOMAP face database consists ofimages (256levels ofgray)ofsize64 64,i.e.,vectors in R4096, whereas the correct intrinsic dimension is only3 (for the vertical, horizontal pause and lightingdirection). The second approach, is anaverage caseapproach (in the spirit of thestatistical mechanics treatment ofhighdimensional systems), thatmodelsfeaturevectorsby arandom ensemble,taken as aset ofrandom vectors with independently identically distributed (i.i.d.) components, and a small but xed fraction of non-zero components.

Similarly the ISOMAP face database consists ofimages (256levels ofgray)ofsize64 64,i.e.,vectors in R4096, whereas the correct intrinsic dimension is only3 (for the vertical, horizontal pause and lightingdirection). The second approach, is anaverage caseapproach (in the spirit of thestatistical mechanics treatment ofhighdimensional systems), thatmodelsfeaturevectorsby arandom ensemble,taken as aset ofrandom vectors with independently identically distributed (i.i.d.) components, and a small but xed fraction of non-zero components.

Recent advancements in vision backbones have significantly improved their performance by simultaneously modeling images' local and global contexts.

We take 1000 training images from CIFAR-10 as a fixed batch, randomly sample the neural architecture for inference, and computevar(µ) of the last BN layer of a NSA and a NSA-i trained givenS = 5000architectures. Inthissection, wecalculate thetestaccuracyof200randomly sampled architectures based onthe vanilla NSA models trained under various spaces. A half of these architectures are seen during trainingwhiletheotherhalfnot.

The predictions provided by different architectures can be further assembled or used to calculate uncertainty estimates, making the prediction model more accurate,robust,andcalibrated.

Previous OOD detection approaches either rely on the feature distance [38], activation abnormality[52],orgradientnorm[27].

By making our data processing source code publiclyavailable, weaim toengage themarine science community toenrich thedata pool andinspire themachine learning community to develop more robust models.

Mostexisting results on Federated Learning (FL) assume the clients have ground-truth labels.