CPSC 522 - Artificial Intelligence 2
Schedule

Which students are presenting on each day: http://wiki.ubc.ca/Course:CPSC522/StudentPresentations2023W1.

• Sept 7 - First Class: AI and Agents. Slides: Introduction, agents, dimensions. Readings: AIFCA Sections 2.1, 2.4, 9.1, 9.2, 9.3
• Sept 12 - controllers, moral machines, probability, belief networks. Readings: AIFCA Sections 2.1, 2.4, 9.1, 9.2, 9.3; rolling average.
• Sept 14 - Discussion paper:
  • Artificial Intelligence - The Revolution Hasn't Happened Yet, Harvard Data Science Review (including commentary and rejoinder), 2019
  independence in belief networks. Demo of reasoning in aipython.org (chapter 9)
• Sept 19 - representing conditional probability, probabilistic inference: recursive conditioning. See AIPython Chapter 9 for Python code implementing belief networks and recursive conditioning (and more).
• Sept 21 - probabilistic inference: variable elimination. temporal models. Discussion papers:
  • Bender, E. M., Gebru, T., McMillan-Major, A., and Shmitchell, S. [2021]. On the dangers of stochastic parrots: Can language models be too big? In Proc. 2021 ACM Conference on Fairness, Accountability, and Transparency,
• Sept 26 - stochastic simulation. Video: sca80a0.avi: Animation of Monte Carlo Localization using laser range finder by Sebastian Thrun. unsupervised learning (EM).
• Sept 28 - Bayesian learning (EM). learning belief networks. Discussion paper:
  • Diederik P Kingma, Max Welling [2013] Auto-Encoding Variational Bayes
• Oct 3 - Causality, Inferring Causality.
• Oct 5 - Missing data, counterfactuals.
Discussion paper:
• Karthika Mohan & Judea Pearl (2021) Graphical Models for Processing Missing Data, Journal of the American Statistical Association, 116:534, 1023-1037, DOI: 10.1080/01621459.2021.1874961 (UBC login required)
• Oct 10 - Preferences, utility and rewards. . Video
• Oct 12 - No Class (makeup Monday)
• Oct 17 - Decision networks. Markov decision processes. Demo of decision networks from aipython.org (chapter 12)
• Oct 24 - Reinforcement Learning and Q-learning, Exploration. Demo of reinforcement learning from aipython.org (chapter 13)
• Oct 26 - Beyond Q-leaning. video. Discussion paper:
  • Kocsis, L., and Szepesvari, C. 2006. Bandit Based Monte-Carlo Planning. In Proceedings of the 17th European Conference on Machine Learning (ECML), 282-293.
• Oct 31 - Multiagent systems representations, algorithms. Demo of multiagent reinforcement learning from aipython.org (chapter 14)
• Nov 2 - mechanism design. Discussion paper:
  • Silver, D., et al. [2017]. Mastering chess and shogi by self-play with a general reinforcement learning algorithm.
• Nov 7 - logic for representation, knowledge graphs. Justin's Family Tree family.pl. Wikidata page on Christine Sinclair described in AIFCA Examples 16.6 and 16.7 and Figure 16.1; also machine readable in RDF.
• Nov 9 - ontologies. Discussion paper:
  • Kemp, C., Perfors, A. and Tenenbaum, J.B. (2007), Learning overhypotheses with hierarchical Bayesian models. Developmental Science, 10: 307-321. Science, Vol. 349, Issue 6245, pp. 273-278, 2015 (you might need to VPN to UBC to get free access)
• Nov 14 - no class (midterm break)
• Nov 9 - relations to random variables. embedding-based methods. Thw demo was for relnCollFilt.py in aipython.org (Ch 15).
• Nov 21 - relational probabilistic models.
• Nov 23 - some specific relational probabilistic models. Discussion paper:
  • Trouillon, T.; Welbl, J.; Riedel, S.; Gaussier, E.; and Bouchard, G. 2016. Complex embeddings for simple link prediction. In ICML, 2071-2080.
• Nov 28 - The social impact of AI; digital economy, values and biases. Human-centered AI.
• Nov 30 - Work, Automation, Transportation.. Sustainability.
• Dec 5 - Discussion paper:
  • Michael Schlichtkrull, Thomas N. Kipf, Peter Bloem, Rianne van den Berg, Ivan Titov, Max Welling [2018]. Modeling relational data with graph convolutional networks. European Semantic Web Conference.
• Dec 7 - AI and Agents. deploying AI. dimensions of AI. the future.

Last updated: 2020-03-15, David Poole