Pruning gets wors with overparametrization Testing their combinatorial method Zhang&Papayan( @stats285 ) find that when adding (unneeded) parameters you end up with more (absolute) number of parameters for the same performance. #ICML2024
bsky.app/profile/lcho...
best #icmi2024 position: 103 datasets that claim to be more diverse, are not. Diversity claims are subjective, political and not tested, instead of claiming, let's measure. But how? arxiv.org/abs/2407.08188 #ML #machinelearning #data #DataScience #NLP #NLProc 🤖