Neural Network Memorization
Tim and Florian discuss the fine line between neural networks memorizing training data and learning generalizable representations. Nicholas highlights the differences in memorization tendencies between language and image models, raising questions about the essence of machine learning processes.In this clip
From this podcast

Machine Learning Street Talk (MLST)
#040 - Adversarial Examples (Dr. Nicholas Carlini, Dr. Wieland Brendel, Florian Tramèr)
Related Questions