Learning with Adversarial Examples
Florian discusses the challenge of determining when a model has enough data to classify accurately. Tim reflects on the potential chaos in data and the creation of adversarial examples. Wieland questions the effectiveness of throwing more data at a problem and emphasizes the importance of understanding the model's inductive bias for more elegant solutions in machine learning.In this clip
From this podcast
Machine Learning Street Talk (MLST)
#040 - Adversarial Examples (Dr. Nicholas Carlini, Dr. Wieland Brendel, Florian Tramèr)
Related Questions