00:00:00 / 00:00:00

Interpreting Deep Neural Networks towards Trustworthiness

By Bin Yu

Appears in collection : A Multiscale tour of Harmonic Analysis and Machine Learning - To Celebrate Stéphane Mallat's 60th birthday

In this talk, I will start by describing our contextual decomposition (CD) method to interpret neural networks, which attributes importance to features and feature interactions for individual predictions. Using CD to interpret DL models in a cosmology problem led us to develop an adaptive wavelet distillation (AWD) interpretation method. AWD is shown to be both outperforming deep neural networks and interpretable in the motivating cosmology problem and an external validating cell biology problem. Finally, I will address the need to quality-control the entire data science life cycle to build any model for trustworthy interpretation.

Information about the video

  • Date of recording 19/04/2023
  • Date of publication 26/04/2023
  • Institution IHES
  • Language English
  • Audience Researchers
  • Format MP4

Domain(s)

Last related questions on MathOverflow

You have to connect your Carmin.tv account with mathoverflow to add question

Ask a question on MathOverflow




Register

  • Bookmark videos
  • Add videos to see later &
    keep your browsing history
  • Comment with the scientific
    community
  • Get notification updates
    for your favorite subjects
Give feedback