Information Bottleneck

1 selectedDifficulty 7-71 unseenView topic

AdvancedNew

0 answered

1 advancedAdapts to your performance

Question 1 of 1

120sadvanced (7/10)compare

The Information Bottleneck (IB) objective min_{encoder} I (X; Z) - β \cdot I (Z; Y) was proposed as an explanation for deep learning generalization: networks allegedly compress the input during a distinct training phase. What is the main critique of this explanation?