The Design of Functional Numerical Software - Dr. Richard Mortier, University of Cambridge

The Design of Functional Numerical Software - Dr. Richard Mortier, University of Cambridge

Alan Turing Institute via YouTube Direct link

Intro

1 of 31

1 of 31

Intro

Class Central Classrooms beta

YouTube playlists curated by Class Central.

Classroom Contents

The Design of Functional Numerical Software - Dr. Richard Mortier, University of Cambridge

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

  1. 1 Intro
  2. 2 Why, What, When, Where?
  3. 3 Motivating Application: Databox
  4. 4 Owl's Architecture
  5. 5 Indexing & Slicing
  6. 6 Core Functor Stack
  7. 7 Advanced Uses of Algorithmic Differentiation
  8. 8 Laziness & Dataflow
  9. 9 Incremental Computation
  10. 10 GPGPU Programming
  11. 11 Expressiveness
  12. 12 Google Inception v3 in 150 LOC
  13. 13 Actor, Parallel and Distributed Processing
  14. 14 Owl & Actor: Neural Network Example
  15. 15 Actor & the Synchronous Parallel Machine
  16. 16 Barrier Synchronisation
  17. 17 Three Main Schemes: A 10,000 Foot View
  18. 18 Simple Analytical Model
  19. 19 Decomposing Synchronous Parallel Machine
  20. 20 Key Insights from System Decomposition
  21. 21 Probabilistic Synchronous Parallel
  22. 22 Sampling Primitive
  23. 23 Adding the Completeness Dimension
  24. 24 Reducing Sample Size
  25. 25 Revisit System Decomposition
  26. 26 Comparing Synchronisation Methods
  27. 27 Step Distribution
  28. 28 Effect of Sample Size
  29. 29 Tightening Bounds
  30. 30 Scalability
  31. 31 Ongoing: Integration with App Development Kit

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.