CPSC 532 - Topics in AI:
Probabilistic graphical models, knowledge graphs and statistical-relational AI
January-April 2021
Here is a tentative schedule. Each week, 2 students
will jointly present the paper. All papers are freely available for us, but
you might need to use a UBC VPN.
- Jan 13 - Graphical Models foundations: Probability,
Independence, graphical models Slides: lect1.pdf
lect2.pdf
- Jan 12 - Assignment 0 due
- Jan 15 - Belief networks and other graphical models (cont). lect3.pdf
- Jan 20 Discussion papers (AI and probability):
-
Zoubin Ghahramani. Probabilistic machine learning and artificial
intelligence. Nature, 521:452-459, 2015, doi
doi:10.1038/nature14541.
- Jan 22 Probabilistic Inference: Exact. Slides: lect7.pdf lect4.pdf
- Jan 27 - Markov Chains, and stochastic simulation. Slides: lect5.pdf
Discussion paper (probabilistic inference):
-
Tian Sang, Paul Beame, and Henry A. Kautz. Performing Bayesian
inference by weighted model counting. In AAAI, 2005.
- Jan 29 - Stochastic simulation (cont). Learning Probabilsitc
Models. Slides: lect6.pdf
ML.pdf.
- Feb 3 Discussion paper (causality)
- Judea Pearl and Elias Bareinboim: External Validity: From
Do-Calculus to Transportability Across Populations. Statistical Science
2014, Vol. 29, No. 4.
- Feb 5 knowledge graphs, maxrix factorization for recommended systems. slides: Poole_talk.pdf
lect4.pdf. Wikidata. AIPython (see Chapter 13).
- Feb 10 Discussion paper (knowledge graphs)
-
Seyed Mehran Kazemi and David Poole, SimplE Embedding for Link
Prediction in Knowledge Graphs, in Proc. Thirty-second Conference on
Neural Information Processing Systems (NeurIPS 2018).
- Feb 12. Learning knowledge graph overview. FB15k relations
(and translations of some
of test set): Sentences2.txt.
Slides
- Feb 12 - see Wiki for February Assignment/Project.
- Feb 24 Discussion paper (everyone reads and 2 people act as
discussion leaders):
-
Query2box: Reasoning over Knowledge Graphs in Vector Space using Box
Embeddings by Hongyu Ren, Weihua Hu, Jure Leskovec (ICLR 2020)
- Mar 3. See Piazza.
- Mar 5: Discussion paper (everyone reads and 2 people act as
discussion leaders):
-
Modeling Relational Data with Graph Convolutional Networks by
Michael Schlichtkrull, Thomas N. Kipf, Peter Bloem, Rianne van den
Berg, Ivan Titov, Max Welling ESWC 2018.
- March 5 - see Wiki for March Assignment/Project. Choose papers.
- Mar 10: Discussion paper (everyone reads and 2 people act as
discussion leaders):
-
End-to-End Differentiable Proving by
Tim Rocktaschel, Sebastian Riedel (NIPS 2017)
- Mar 17 - recruiting talk
- Mar 19 - relational probabilistic models. Slides:
lect3.pdf
- Mar 24 - (exact) lifted inference
- March 26
-
Mathias Niepert and Guy Van den Broeck,
lifted MCMC
2021.
- March 31
- April 9
Last updated: 2021-03-22 David Poole