3 Explaining Deep Neural Network Models

XAI Course Notes

Rigorously understanding how Deep Learning models function may allow us to steer and control their behavior and avoid unwanted consequences. In addition, today’s highly capable AI systems could be useful for studying intelligence. Can we reverse engineer neural nets from their weights, similar to how one might reverse engineer a binary compiled program? Can we gain insights into neural computation similar to how neuroscientists study the brain?

explainable AI

XAI

machine learning

data science

contrafactuals

casual inference

Series Poseter

Session Video

not available!

Reuse

CC SA BY-NC-ND

Citation

BibTeX citation:

@online{bochman2023,
  author = {Bochman, Oren},
  title = {3 {Explaining} {Deep} {Neural} {Network} {Models}},
  date = {2023-03-19},
  url = {https://orenbochman.github.io/notes/XAI/l03/},
  langid = {en}
}

For attribution, please cite this work as:

Bochman, Oren. 2023. “3 Explaining Deep Neural Network Models.” March 19, 2023. https://orenbochman.github.io/notes/XAI/l03/.