I-I Reading Group

I-I (iDEA-iSAIL) group is a statistical learning and data mining reading group at UIUC, coordinated by Prof. Hanghang Tong and Prof. Jingrui He. The main purpose of this group is to educate and inform its members of the recent advances of machine learning and data mining.

Regular Meetting Time: {Wed 11 am} at {Thomas M. Siebel Center, 4102}.

Unless otherwise notified, our regular weekly meeting for Fall 2019 is Wed 11-12am, SC 4102. If you would like to present in an upcoming meeting, please email dzhou21/lecheng4 [at] illinois [dot] edu or submit a pull request and add to the table!

Schedule for 2019 Fall:

Dates	Presenters	Topics	Materials
Sep 4, 2019	Hansheng Ren	Kick-Off Meeting	-
Sep 11, 2019	Xue Hu & Dongqi Fu	Data Poisoning Attacks
Sep 18, 2019	Dawei Zhou & Lecheng Zheng	Recent Advances of Transformer Machine	-
Sep 25, 2019	Yao Zhou & Jun Wu	TBD	-
Oct 2, 2019	MAA-members	Active Interpretation of Disparate Alternatives (AIDA)	-
Oct 9, 2019	Zhe Xu	-	-
Oct 16, 2019	Si Zhang & Qinghai Zhou	-	-
Oct 23, 2019	Yu Wang & Ziye Zhu	Language Model in NLP	-
Nov 6, 2019	Lihui Liu	-	-
Nov 13, 2019	-	-	-
Nov 20, 2019	-	-	-
Nov 27, 2019	NA	Thanksgiving	-
Dec 4, 2019	-	-	-

Recommended Flows

Introduce 1~2 Research Papers:

• 20 mins: Introduction & Background (Motivation examples, literature review)
• 10 min: Problem Description (Give a formal definition of the studied problems)
• 30 min: Brainstorm Discussion (Propose potential approaches based on your knowledge)
• 30 min: Algorithm (Description of the algorithms in the papers)
• 30 min: Critical Discussion (Pros & Cons of your ideas and the existing one)

Survey a Research Topic

• 20 mins: Introduction & Background (Motivation examples, literature review)
• 20 min: Problem/Subproblems Description (Give a formal definition of the studied problems)
• 60 min: Review (High-level discussion of the existing work)
• 20 min: Conclusion & Future Direction

Covered topics/papers in the past:

Generative Deep Learning:

• Martín Arjovsky, Soumith Chintala, Léon Bottou: Wasserstein Generative Adversarial Networks. ICML 2017: 214-223 
• Gulrajani, Faruk Ahmed, Martín Arjovsky, Vincent Dumoulin, Aaron C. Courville: Improved Training of Wasserstein GANs. NIPS 2017: 5767-5777 
• You, Jiaxuan, et al. "Graphrnn: Generating realistic graphs with deep auto-regressive models." arXiv preprint arXiv:1802.08773 (2018). 
• Aleksandar Bojchevski, Oleksandr Shchur, Daniel Zügner, Stephan Günnemann: NetGAN: Generating Graphs via Random Walks. ICML 2018: 609-618 

Robustness:

• Eric Wong, J. Zico Kolter: Provable Defenses against Adversarial Examples via the Convex Outer Adversarial Polytope. ICML 2018: 5283-5292.  

Meta Learning:

• Chelsea Finn, Pieter Abbeel, Sergey Levine: Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks. ICML 2017: 1126-1135. 

Fairness Learning:

• Tolga Bolukbasi, Kai-Wei Chang, James Y. Zou, Venkatesh Saligrama, Adam Tauman Kalai: Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings. NIPS 2016: 4349-4357.  
• Richard S. Zemel, Yu Wu, Kevin Swersky, Toniann Pitassi, Cynthia Dwork: Learning Fair Representations. ICML (3) 2013: 325-333.  

Adversarial Attacks:

• Hanjun Dai, Hui Li, Tian Tian, Xin Huang, Lin Wang, Jun Zhu, Le Song: Adversarial Attack on Graph Structured Data. ICML 2018: 1123-1132 . 
• Daniel Zügner, Amir Akbarnejad, Stephan Günnemann: Adversarial Attacks on Neural Networks for Graph Data. KDD 2018: 2847-2856. 
• Guanhong Tao, Shiqing Ma, Yingqi Liu, Xiangyu Zhang: Attacks Meet Interpretability: Attribute-steered Detection of Adversarial Samples. NeurIPS 2018: 7728-7739 

Click to see what we have covered in each semester