Logical Neural Networks: Unifying Statistical and Symbolic AI

Logical Neural Networks: Unifying Statistical and Symbolic AI

Georgia Tech Research via YouTube Direct link

Summary

27 of 27

27 of 27

Summary

Class Central Classrooms beta

YouTube videos curated by Class Central.

Classroom Contents

Logical Neural Networks: Unifying Statistical and Symbolic AI

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

  1. 1 Intro
  2. 2 Logical Neural Networks
  3. 3 Neuro-symbolic methods so far
  4. 4 Neuro-symbolic methods: another category
  5. 5 Original (McCulloch and Pitts 1943) neuron as logic gate
  6. 6 Weighted neuron (perceptron, 1958) as logic gate
  7. 7 Differentiable neuron (MLPs, deep learning) as logic gate
  8. 8 b. Constrained differentiable neuron (LNN) as logic gate
  9. 9 a. Neuron (LNN) as real-valued logic gate
  10. 10 a. Neural network inference as logical reasoning
  11. 11 a. Data and learning
  12. 12 7b. Data and learning
  13. 13 Equivalence between neural networks and symbolic logic
  14. 14 Comparison to other common neuro-symbolic ideas
  15. 15 Use case: Knowledge base question answering (KBQA)
  16. 16 KBQA: Why it challenges default Al (end-to-end deep learning)
  17. 17 KBQA: an approach via understanding
  18. 18 Making the model & inference process human-understandable
  19. 19 Learning to reason
  20. 20 Logical rule induction (ILP)
  21. 21 Optimization/learning
  22. 22 Reinforcement learning
  23. 23 Policy induction via rule learning
  24. 24 AGI: Bengio-Marcus Desiderata
  25. 25 Ongoing directions
  26. 26 Philosophical shift
  27. 27 Summary

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.