UAI Workshops Program

Monday, July 15th

---

Workshops

1. Bayes Application Workshops

2. Approaches to Causal Structure Learning Workshop

3. New Challenges in E-Commerce Recommendations

---

Bayes Application Workshops (Room: Lake Washington A)

8:30am – Combined Applications Workshop Introductions

Application Workshop I: Big Data meets Complex Models

9:00 - 10:15

Session 1: Extracting information about individuals from complex data

1. Identifying Learning Trajectories in an Educational Video Game – Kerr and Chung
2. Using information to debug Complex Systems – Almond, Kim, Shute, & Ventura
3. Transforming Personal Artifacts into Probabilistic Narratives – Rafatirad and Laskey

15' Open Discussion of Session 1

10:15 - 10:30 Coffee Break

10:30 - 11:45

Session 2: Combining Learning and Expert Opinion

1. Bayesian Supervised Dictionary Learning – Babagholami-Mohamadabadi, Jourabloo, Zolfaghari, & Manzuir-Shalmani
2. Learning Parameters by Prediction Markets and Kelly Rule for Graphical Models – Sun, Hanson, Twardy, & Laskey
3. Predicting Latent Variables with Knowledge and Data: A Case Study in Trauma Care – Yet, Marsh, Perkins, Tai, & Fenton

15' Open Discussion of Session 2

11:45 - 13:00 Break (lunch on your own)

Application Workshop 2: Spatial, Temporal, and Network Models

13:00 - 14:30

Session 3

1. An Object-oriented Spatial and Temporal Bayesian Network for Managing Willows in an American Heritage River Catchment – Wilkinson, Chee, Nicholson, & Quintana-Ascencio
2. Product Trees for Gaussian Process Covariance in Sublinear Time – Moore & Russell
3. Latent Topic Analysis for Predicting Group Purchasing Behavior on the Social Web – Sun, Yeh, Mengshoel, & Griss

15' Open Discussion of Session 3

14:30 - 14:45 Coffee Break

14:45 - 15:45

Session 4

1. A lightweight inference method for image classification – Agosta and Pillai
2. Exploring Multiple Dimensions of Parallelism in Junction Tree Message Passing – Zheng and Mengshoel

10' Open Discussion of Session 3

15:45 - 16:00 Coffee Break

16:00 - 17:00 Combined Applications Workshop Wrap-up and Planning Discussion

---

Approaches to Causal Structure Learning Workshop (Room: Lake Chelan)

8:45 - 10:00 Session 1 (75 minutes)

• 8:45 - 9:20 Invited talk - David Heckerman
• 9:20 - 9:55 Discovering Cyclic Causal Models with Latent Variables: A General SAT-Based Procedure - Hyttinen, Hoyer, Eberhardt, Järvisalo

10:00 - 10:30 Coffee Break

10:30 - 12:15 Session 2 (105 minutes)

• 10:30 - 11:05 Reasoning about Independence in Probabilistic Models of Relational Data - Maier, Marazopoulou, Jensen
• 11:05 - 11:40 A Sound and Complete Algorithm for Learning Causal Models from Relational Data - Maier, Marazopoulou, Arbour, Jensen
• 11:40 - 12:15 Single World Intervention Graphs - Richardson and Robins

12:15 - 14:00 Break (lunch on your own)

14:00 - 15:25 Session 3 (85 minutes)

• 14:00 - 14:50 Invited Talk - Joris Mooij
• 14:50 - 15:25 Identifiability of binary directed graphical models with hidden variables - Allman, Rhodes, Stanghellini and Valtorta

15:25 - 15:35 Poster spotlights

15:35 - 16:50 Posters (and coffee from 16:00)

16:50 - 18:00 Session 4 (70 minutes)

• 16:50 - 17:25 Maximum Likelihood estimation of structural nested logistic model with an instrumental variable - Matsouaka and Tchetgen
• 17:25 - 18:00 A finite population test of the sharp null hypothesis for Compliers - Loh and Richardson

---

New Challenges in E-Commerce Recommendations (Room: Lake Coeur d'Alene)

8:30 - 9:30 Alex Smola

9:30 - 10:00 Coffee Break

Oral Presentation 1

• 10:00 - 10:30 Ratings Re-specification for Rank Ordered Recommendations - Oluwasanmi Koyejo, Sreangsu Archaryya, Joydeep Ghosh
• 10:30 - 11:00 On the Evaluation of Rating-Prediction and Ranking - Harald Steck
• 11:00 - 11:30 Exploiting Similarity to Optimize Recommendations Lists from Click Feedback - Hastagiri Prakash Vanchinathan, Isidor Nikolic, Fabio De Bona, Andreas Krause
• 11:30 - 12:15 Collaborative filtering Graphlab - Danny Bickson

12:15 - 14:00 Break (lunch on your own)

14:00 - 15:00 Recommendations at Netflix - Carlos Gomez Uribe (Netflix)

15:00 - 15:15 Coffee Break

Oral Presentation 2

• 15:15 - 15:45 Lei Tang (Walmart Labs)
• 15:45 - 16:15 Vaclav Petricek (eHarmony)
• 16:15 - 16:45 Bo Long A ds Recommendation and Large-Scale Machine Learning at LinkedIn

16:45 - 17:00 Panel Discussion