Upgrade to Pro — share decks privately, control downloads, hide ads and more …

文献紹介 / Structure-based Knowledge Tracing: An In...

Atom
May 06, 2021
0
76

文献紹介 / Structure-based Knowledge Tracing: An Influence Propagation View

Atom

May 06, 2021
Tweet

More Decks by Atom

See All by Atom
文献紹介 / Knowledge Tracing with GNN
roraidolaurent
0
76
文献紹介 / Non-Intrusive Parametric Reduced Order Models withHigh-Dimensional Inputs via Gradient-Free Active Subspace
roraidolaurent
0
45
ニューラルネットワークのベイズ推論 / Bayesian inference of neural networks
roraidolaurent
1
2.7k
Graph Convolutional Networks
roraidolaurent
0
200
文献紹介 / A Probabilistic Annotation Model for Crowdsourcing Coreference
roraidolaurent
0
54
文献紹介Deep Temporal-Recurrent-Replicated-Softmax for Topical Trends over Time
roraidolaurent
0
83
文献紹介/ Bayesian Learning for Neural Dependency Parsing
roraidolaurent
0
88
ポッキー数列の加法定理 / Pocky number additon theorem
roraidolaurent
0
190
Scalable Bayesian Learning of Recurrent Neural Networks for Language Modeling
roraidolaurent
1
120

Featured

See All Featured
How GitHub (no longer) Works
holman
310
140k
Making Projects Easy
brettharned
115
5.9k
The Art of Delivering Value - GDevCon NA Keynote
reverentgeek
8
900
Raft: Consensus for Rubyists
vanstee
136
6.6k
Scaling GitHub
holman
458
140k
Building Better People: How to give real-time feedback that sticks.
wjessup
364
19k
Facilitating Awesome Meetings
lara
50
6.1k
Evolution of real-time – Irina Nazarova, EuRuKo, 2024
irinanazarova
4
370
Why You Should Never Use an ORM
jnunemaker
PRO
54
9.1k
A Modern Web Designer's Workflow
chriscoyier
693
190k
For a Future-Friendly Web
brad_frost
175
9.4k
Statistics for Hackers
jakevdp
796
220k

Transcript

  1. None

  2. 2

  3. 3

  4. 4

  5. 5

  6. 6

  7. 7

  8. 8

  9. 9

  10. 10 𝒙𝑡 = 𝑒𝑡 = 0に回答して不正解? 𝑒𝑡 = 0に回答して正解? 𝑒𝑡

    = 1に回答して不正解? ⋮ 関係r ごとに埋め込み 関係r に関する一時的に用いる特徴量 注）ℎ𝑖 𝑡 ← ℎ 𝑖 𝑡,𝑇と更新されない．

  11. 11 𝑝𝑎𝑟𝑡𝑖𝑗 𝑟

  12. 12

  13. 13

  14. 14

  15. 15

  16. 16

  17. 17

  18. 18

  19. 19

  20. 20

  21. 21

  22. 22

  23. 23 1ステップのみの学習で得られた 影響ベクトル 𝐽𝑖 ∈ ℝ+𝑁として 概念間の影響ベクトルのコサイン類似度が 0.5以上のものにエッジを引いてクラスタリング

  24. 24

  25. 25