Scaling Laws in Jet Classification
This addresses the problem of classifier selection in collider physics by revealing scaling behaviors, though it is incremental as it extends known scaling laws to a new domain.
The study demonstrated that six physically-motivated classifiers for top versus QCD jet classification exhibit power-law scaling of test loss with training set size, with distinct indices, showing that optimal classifier choice can change as dataset size increases.
We demonstrate the emergence of scaling laws in the benchmark top versus QCD jet classification problem in collider physics. Six distinct physically-motivated classifiers exhibit power-law scaling of the binary cross-entropy test loss as a function of training set size, with distinct power law indices. This result highlights the importance of comparing classifiers as a function of dataset size rather than for a fixed training set, as the optimal classifier may change considerably as the dataset is scaled up. We speculate on the interpretation of our results in terms of previous models of scaling laws observed in natural language and image datasets.