Dataset Groups Activity Stream Deep pNML The pNML scheme outperforms the ERM in the accuracy and log-loss sense, it is more robust for noisy trainset and has better performance for adversarial samples. BibTex: @dataset{Koby_Bibas_and_Yaniv_Fogel_and_Meir_Feder_2024, abstract = {The pNML scheme outperforms the ERM in the accuracy and log-loss sense, it is more robust for noisy trainset and has better performance for adversarial samples.}, author = {Koby Bibas and Yaniv Fogel and Meir Feder}, doi = {10.57702/l37quzgj}, institution = {No Organization}, keyword = {'Deep Neural Networks', 'ERM', 'Universal Learning', 'pNML'}, month = {dec}, publisher = {TIB}, title = {Deep pNML}, url = {https://service.tib.eu/ldmservice/dataset/deep-pnml}, year = {2024} }