Insights on Gradient-Based Algorithms in High-Dimensional Learning

Insights on Gradient-Based Algorithms in High-Dimensional Learning

Simons Institute via YouTube Direct link

WHY THIS MODEL?

5 of 29

5 of 29

WHY THIS MODEL?

Class Central Classrooms beta

YouTube videos curated by Class Central.

Classroom Contents

Insights on Gradient-Based Algorithms in High-Dimensional Learning

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

  1. 1 Intro
  2. 2 WORKHORSE OF MACHINE LEARNIN
  3. 3 IN DEEP LEARNING
  4. 4 STRATEGY
  5. 5 WHY THIS MODEL?
  6. 6 ESTIMATORS
  7. 7 GRADIENT-BASED ALGORITHMS
  8. 8 DYNAMICAL MEAN FIELD THEORY
  9. 9 LANGEVIN STATE EVOLUTION (NUMERICAL SOLUTION)
  10. 10 LANGEVIN PHASE DIAGRAM
  11. 11 GRADIENT-FLOW PHASE DIAGRAM
  12. 12 POPULAR "EXPLANATION"
  13. 13 SPURIOUS MINIMA DO NOT NECESSARILY CAUSE GF TO FAIL
  14. 14 WHAT IS GOING ON?
  15. 15 TRANSITION RECIPE
  16. 16 TRANSITION CONJECTUR
  17. 17 LANDSCAPE ANALYSIS
  18. 18 CONCLUSION ON SPIKED MATRIX-TENSOR MODEL
  19. 19 TEACHER-NEURAL SETTING
  20. 20 TEACHER STUDENT PERCEPTRON
  21. 21 PHASE RETRIEVAL: OPTIMAL SOLUTION
  22. 22 GRADIENT DESCENT FOR PHASE RETRIEVAI
  23. 23 PERFORMANCE OF GRADIENT DESCENT
  24. 24 GRADIENT DESCENT NUMERICALLY
  25. 25 TOWARDS A THEORY
  26. 26 OVER-PARAMETRISED LANDSPACE
  27. 27 STOCHASTIC GRADIENT DESCENT
  28. 28 DYNAMICAL MEAN-FIELD THEOR Mignaco, Urbani, Krzakala, LZ, 2006.06098
  29. 29 DMFT FOLLOWS THE WHOLE TRAJECTORY

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.