How Can You Trust Machine Learning?

How Can You Trust Machine Learning?

Stanford HAI via YouTube Direct link

Oversensitivity in image classification

13 of 23

13 of 23

Oversensitivity in image classification

Class Central Classrooms beta

YouTube videos curated by Class Central.

Classroom Contents

How Can You Trust Machine Learning?

Automatically move to the next video in the Classroom when playback concludes

  1. 1 Intro
  2. 2 Math Myth of ML (circa 2008)
  3. 3 spaces between the Math
  4. 4 trust for whom?
  5. 5 Train a neural network to predict wolf v. husky
  6. 6 Explanations for neural network prediction
  7. 7 Accuracy vs Interpretability
  8. 8 Explaining predictions
  9. 9 Explaining prediction of Inception Neural Network
  10. 10 Anchors for Visual Question Answering
  11. 11 Type 1 Diabetes Management
  12. 12 Standard Intervention
  13. 13 Oversensitivity in image classification
  14. 14 Beyond Test-Set Accuracy
  15. 15 Closing the Loop with Simple Data Augmentation
  16. 16 Checklist: Test Linguistic Capabilities of Model
  17. 17 Checklist: Categories of Tests
  18. 18 Addressing Challenge of Test Creation
  19. 19 User Study: Quora Question Pairs (n=18, 2 hours)
  20. 20 Minding the Gap
  21. 21 Adaptive Loss Alignment (ALA)
  22. 22 And this gap is increasing with foundation models...
  23. 23 Optimizing for multiple metrics

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.