Efficient Hyperparameter Tuning
Gábor emphasizes the importance of efficient hyperparameter tuning on varying data set sizes, cautioning against transferring values across different datasets. Matt highlights the significant drop in performance when tuning on smaller datasets and applying to larger ones, stressing the dataset-specific nature of hyperparameters.In this clip
From this podcast
NLP Highlights
40 - On the State of the Art of Evaluation in Neural Language Models, with Gábor Melis
Related Questions