Syllabus and Textbooks

The required reading will be mainly from select conference/journal articles (listed below for each lecture), which are all available online, as well as a few chapters of the following textbook:

• [GRL] Graph Representation Learning
  William L. Hamilton

Syllabus

Prior to every class, we provide a list of papers - select research papers relevant to a specific lecture. Students are encouraged to read these papers before the class. See further details in Policies section. GRL indicates the above text. Slides, lecture ntoes and other materials will be available on Blackboard.

Week 1: Introduction and Basics

• Ch.01 Introduction [GRL]

• Ch.22 Elementary Graph Algorithms [CLRS]

• Ch.24 Single Source Shortest Paths [CLRS]

Week 2: Graph Properties and Features I

• Ch.02 Background and Traditional Approaches [GRL]

• Global connectivity and multilinguals in the Twitter network. Hale, S.A. SIGCHI’14

• Searching for superspreaders of information in real-world social media. Pei, S., et al. Scientific reports’14.

• Assignment 1 (out), Project Proposal (out)

Week 3: Graph Properties and Features II

• Ch.02 Background and Traditional Approaches [GRL]

• Graph structure in the web. Broder, A., et al. Computer networks’00.

• Structure and tie strengths in mobile communication networks. Onnela, J.P., et al. PNAS’07.

Week 4: Node Embedding I

• Distributed representations of words and phrases and their compositionality. Mikolov, T., et al. NIPS’13.

• Glove: global vectors for word representation. Pennington, J., et al. EMNLP’14.

• Evaluation methods for unsupervised word embeddings. Schnabel, T., et al. EMNLP’15.

• Tutorial: Distributional semantics.

• Assignment 1 (in), Assignment 2 (out)

Week 5: Node Embedding II

• Ch.03 Neighborhood Reconstruction Method [GRL]

• Node2vec: scalable feature learning for networks. Grover, A. and Leskovec, J. SIGKDD’16.

• Deepwalk: online learning of social representations. Perozzi, B., et al. SIGKDD’14.

• Course Lecture: Graph Representation Learning. Leskovec J. 2019.

• Assignment 2 (in), Assignment 3 (out)

Week 6: Node Embedding III

• Retrofitting word vectors to semantic lexicons. Faruqui, M., et al. NAACL’15.

• AI2 talk: Beyond the Distributional Hypothesis: Learning Better Word Representations. Manaal F. 2016.

• Assignment 3 (in)

Week 7: Exam

• Midterm exam
  

• See details on Blackboard.

Week 8: Graph Neural Networks I

• Ch.05 The Graph Neural Network Model [GRL]

• Random walk graph neural networks. Giannis, N. et al. NeurIPS’20

• Inductive representation learning on large graphs. Hamilton, W., et al. NIPS’17.

• Semi-supervised classification with graph convolutional networks. Kipf, T.N., et al. ICLR’17.

• Assignment 4 (out)

Week 9: Projects

• Proposal presentations
  

• See details on Blackboard.

• Project Proposal (in), Project Report (out).

Week 10: Graph Neural Networks II

• Strategies for pre-training graph neural networks. Hu, Weihua, et al. ICLR’20.

• Learning convolutional neural networks for graphs. Niepert, M., et al. ICML’16.

• Assignment 5 (out)

Week 11: Graph Neural Networks III

• Graph convolutional neural networks for web-scale recommender systems. Ying, R., et al. KDD’18.

• Tutorial: GraphSAGE.

• Tutorial: StellarGraph.

Week 12: Meta Learning with Graphs I

• Neural self-training through spaced repetition. Amiri H. NAACL’19.

• Mentornet: Learning data-driven curriculum for very DNNs on corrupted labels. Lu, J., et al. ICML’18.

• Spaced repetition for efficient and effective training of neural networks. Amiri H., et al. EMNLP’17.

• Self-paced curriculum learning. Lu, J., et al. AAAI’15.

• Curriculum learning. Yoshua, B. et al. ICML’09.

• Tutorial: Leitner system.

• Assignment 4 (in)

Week 13: Thanksgiving Recess

• no class.

Week 14: Meta Learning with Graphs II

• CurGraph: curriculum learning for graph classification. Wang, Yiwei, et al. WWW’21.

• Curriculum learning by dynamic instance hardness. Zhou, T., et al. NeurIPS’20.

• SuperLoss: a generic loss for robust curriculum learning. Castells, et al. NeurIPS’20.

• Multi-Modal curriculum learning over graphs. Chen, G., et al. TIST’19

• Assignment 5 (in)

Week 15: Projects

• Final project presentations.

• See details on Blackboard.

• Project Report (in).