Knowledge Is Embedded in Language Neural Networks but Can They Reason?

Knowledge Is Embedded in Language Neural Networks but Can They Reason?

Simons Institute via YouTube Direct link

GPT-2 language model (cherry-picked) output

6 of 16

6 of 16

GPT-2 language model (cherry-picked) output

Class Central Classrooms beta

YouTube videos curated by Class Central.

Classroom Contents

Knowledge Is Embedded in Language Neural Networks but Can They Reason?

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

  1. 1 Intro
  2. 2 Tree-structured models
  3. 3 Language Modeling
  4. 4 LMs in The Dark Ages: n-gram models
  5. 5 Enlightenment era neural language models (NLMs)
  6. 6 GPT-2 language model (cherry-picked) output
  7. 7 Transformer models
  8. 8 Classic Word Vec (Mikolov et al. 2013)
  9. 9 Self-attention in (masked) sequence model
  10. 10 Good systems are great, but still basic NLU errors
  11. 11 What is Reasoning? Bottou 2011
  12. 12 Appropriate structural priors
  13. 13 Compositional reasoning tree
  14. 14 A 2020s Research Direction
  15. 15 A Neural State Machine
  16. 16 NSM accuracy on GQA

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.