StarAI 2020

Format

StarAI will be a one day workshop with short paper presentations, a poster session, and three invited speakers.

• Guy Van den Broeck (UCLA) 
• Lise Getoor (UC Santa Cruz) 
• Yejin Choi (University of Washington & Allen Institute for AI) 

Submissions

Authors should submit either a full paper reporting on novel technical contributions or work in progress (AAAI style, up to 7 pages excluding references), a short position paper (AAAI style, up to 2 pages excluding references), or an already published work (verbatim, no page limit, citing original work) in PDF format via EasyChair. All submitted papers will be carefully peer-reviewed by multiple reviewers and low-quality or off-topic papers will be rejected. Accepted papers will be presented as a short talk and poster.

Important Dates

• Paper Submission: November 15 18, 2019
• Notification of Acceptance: December 6, 2019
• Camera-Ready Papers: January 15, 2020
• Date of Workshop: February 7, 2020

Morning

•  8:30 a.m.: Welcome and introduction
•  8:35 a.m.: Invited talk Guy Van den Broeck
  
  • Title: Querying advanced probabilistic models: from relational embeddings to probabilistic programs

    Abstract:This talk will go over two recent developments in modeling and querying highly complex probability distributions. First, I will talk about running symbolic queries on relational embedding models, which live in vector space, as if they were probabilistic databases. Surprisingly, this can be done in a way that is sound, exact, accurate, and tractable. Our approach, called TractOR, overcomes some key theoretical limitations of more traditional probabilistic databases, while vastly extending the applicability of relational embedding models to go far beyond link prediction. Second, I will talk about an even richer class of distributions: discrete probabilistic programs. We introduce a novel probabilistic programming language, called Dice. It combines key insights from the AI and programming languages community to open up the black box of probabilistic program inference. This approach results in a highly efficient inference strategy that scales to probabilistic programs with tens of thousands of lines of code.
    (slides)

• 9:35 a.m.: Poster Spotlights (90 seconds each)
• 10:10 a.m.: Coffee break & Poster Session (#1-21)
• 11:30 a.m.: Invited talk Lise Getoor
  
  • Title: Probabilistic Soft Logic: A Scalable, Declarative Approach to Structured Prediction from Noisy Data

    Abstract: A fundamental challenge in developing impactful artificial intelligence technologies is balancing the ability to model rich, structured domains with the ability to scale to big data. Many important problem areas are both richly structured and large scale, including computational social science problems, knowledge graphs and more. In this talk, I will describe Probabilistic Soft Logic (PSL), a declarative probabilistic programming language that is able to both capture rich structure and scale to big data. The mathematical framework upon which PSL is based, hinge-loss Markov random fields (HL-MRFs), is a new kind of probabilistic graphical model that generalizes three different approaches to inference. I show that all three views lead to the same inference objective, and that this inference objective is convex, leading to highly efficient inference algorithms. These inference algorithms can scale further via the use of smart grounding strategies, lifting and more. Along the way, I’ll describe results in a variety of different domains.
    (slides)

• 12:30 p.m.: Lunch break (on your own)

Afternoon

• 2:00 p.m.: Invited talk Yejin Choi
  
  • Title: Neuro-Symbolic Commonsense Intelligence: Cracking the Longstanding Challenge in AI

    Abstract: Despite considerable advances in deep learning, AI remains to be narrow and brittle. One fundamental limitation comes from its lack of commonsense intelligence: reasoning about everyday situations and events, which in turn, requires knowledge about how the physical and social world works.
    In this talk, I will share some of our recent efforts that attempt to crack commonsense intelligence by neural representation learning of commonsense knowledge and reasoning. First, I will introduce ATOMIC, the atlas of relational commonsense knowledge, organized as a graph of 877k if-then rules (e.g., "if X pays Y a compliment, then Y will likely return the compliment”). Next, I will introduce COMET, our neural commonsense models that can learn from and generalize well beyond the knowledge provided in the symbolic ATOMIC knowledge graph to successfully reason about previously unseen events. Finally, I will present a suite of commonsense AI benchmarks ranging from adductive and counterfactual reasoning to visual commonsense reasoning. I will conclude the talk by discussing major open research questions, including the importance of algorithmic solutions to reduce incidental biases in data that can lead to overestimation of true AI capabilities.
    (slides)

• 3:00 p.m.: Poster Spotlights (90 seconds each)
• 3:40 p.m.: Coffee break & Poster Session (#22-46)
• 5:00 p.m.: Panel and wrapup

1. Vaishak Belle
   Abstracting Probabilistic Models: A Logical Perspective
2. Anton Fuxjaeger and Vaishak Belle
   Scaling up Probabilistic Inference in Linear and Non-Linear Hybrid Domains by Leveraging Knowledge Compilation
3. Somak Aditya and Atanu Sinha
   Uncovering Relations for Marketing Knowledge Representation
4. Tao Li, Vivek Gupta, Maitrey Mehta and Vivek Srikumar
   A Logic-Driven Framework for Consistency of Neural Models
5. Efthymia Tsamoura, Victor Gutierrez-Basulto and Angelika Kimmig
   Beyond the Grounding Bottleneck: Datalog Techniques for Inference in Probabilistic Logic Programs
6. Michael Varley and Vaishak Belle
   Implementing Fairness with Tractable Probabilistic Models
7. Steven Holtzen, Todd Millstein and Guy Van den Broeck
   Generating and Sampling Orbits for Lifted Probabilistic Inference
8. Ying Jin, Weilin Fu, Jian Kang, Jiadong Guo and Jian Guo
   Bayesian Symbolic Regression
9. Jonathan Brophy and Daniel Lowd
   EGGS: A Flexible Approach to Relational Modeling of Social Network Spam
10. Marcel Gehrke, Ralf Möller and Tanya Braun
    Taming Reasoning in Temporal Probabilistic Relational Models
11. Ondřej Kuželka and Yuyi Wang
    Domain-Liftability of Relational Marginal Polytopes
12. Yuqiao Chen, Yibo Yang, Sriraam Natarajan and Nicholas Ruozzi
    Lifted Hybrid Variational Inference
13. Sima Behpour
    Active Learning in Video Tracking
14. Varun Embar, Sriram Srinivasan and Lise Getoor
    Estimating Aggregate Properties In Relational Networks With Unobserved Data
15. Bahare Fatemi, Perouz Taslakian, David Vazquez and David Poole
    Knowledge Hypergraphs: Prediction Beyond Binary Relations
16. Alexander Hayes
    srlearn: A Python Library for Gradient-Boosted Statistical Relational Models
17. Pegah Jandaghi and Jay Pujara
    Human-like Time Series Summaries via Trend Utility Estimation
18. Junkang Li, Solene Thepaut and Veronique Ventos
    Reducing incompleteness in the game of Bridge using PLP
19. Gaurav Sinha, Ayush Chauhan, Aurghya Maiti, Naman Poddar and Pulkit Goel
    Dis-entangling Mixture of Interventions on a Causal Bayesian Network UsingAggregate Observation
20. Bishwamittra Ghosh and Kuldeep S Meel
    IMLI: An Incremental Framework for MaxSAT-Based Learning of Interpretable Classification Rules
21. Bassem Makni, Ibrahim Abdelaziz and Jim Hendler
    Explainable Deep RDFS Reasoner
22. Shubham Sharma, Subhajit Roy, Kuldeep S. Meel and Mate Soos
    GANAK: A Scalable Probabilistic Exact Model Counter
23. Yaqi Xie, Ziwei Xu, Mohan S. Kankanhalli, Kuldeep S. Meel and Harold Soh
    Embedding Symbolic Knowledge into Deep Networks
24. Vaishak Belle and Luc De Raedt
    Semiring Programming: A Declarative Framework for Generalized Sum Product Problems
25. Vaishak Belle
    SMT + ILP
26. Teodora Baluta, Shiqi Shen, Shweta Shinde, Kuldeep S. Meel and Prateek Saxena
    Quantitative Verification of Neural Networks and Its Security Applications
27. Yuhao Wang, Vlado Menkovski, Hao Wang, Xin Du and Mykola Pechenizkiy
    Causal Discovery from Incomplete Data: A Deep Learning Approach
28. Pedro Zuidberg Dos Martires and Samuel Kolb
    Monte Carlo Anti-Differentiation for Approximate Weighted Model Integration
29. Serena Booth, Ankit Shah, Yilun Zhou and Julie Shah
    Sampling Prediction-Matching Examples in Neural Networks: A Probabilistic Programming Approach
30. David Arbour, Ryan Rossi, Somdeb Sarkhel and Nesreen Ahmed
    Causal Inference for Graph-based Relational Time Series
31. Michael Skinner, Lakshmi Raman, Neel Shah, Abdelaziz Farhat and Sriraam Natarajan
    A Preliminary Approach for Learning Relational Policies for the Management of Critically Ill Children
32. Johan Pauwels, Gyorgy Fazekas and Mark B. Sandler
    A Critical Look at the Applicability of Markov Logic Networks for Music Signal Analysis
33. Tanya Braun and Ralf Möller
    Exploring Unknown Universes in Probabilistic Relational Models
34. Devendra Dhami, Siwen Yan, Gautam Kunapuli and Sriraam Natarajan
    Non-Parametric Learning of Gaifman Models
35. Adrian Phoulady, Ole-Christoffer Granmo, Saeed R. Gorji and Hady Ahmady Phoulady
    The Weighted Tsetlin Machine: Compressed Representations with Weighted Clauses
36. Navdeep Kaur, Gautam Kunapuli and Sriraam Natarajan
    Non-Parameteric Learning of Lifted Restricted Boltzmann Machines
37. Mayukh Das, Nandini Ramanan, Janardhan Rao Doppa and Sriraam Natarajan
    One-Shot Induction of Generalized Logical Concepts via Human Guidance
38. Matthew Wai Heng Chung and Hegler Tissot
    Evaluating the Effectiveness of Margin Parameter when Learning Knowledge Embedding Representation for Domain-specific Multi-relational Categorized Data
39. Nimar Arora, Nazanin Tehrani, Kinjal Shah, Lily Li, Narges Torabi, Michael Tingley, David Noursi, Sepehr Masouleh, Eric Lippert and Erik Meijer
    Newtonian Monte Carlo: single-site MCMC meets second-order gradient methods
40. Tal Friedman and Guy Van den Broeck
    Towards a Tractable Relational Embedding Model
41. Saket Dingliwal, Ronak Agarwal, Happy Mittal and Parag Singla
    Advances in Symmetry Breaking for SAT Modulo Theories
42. Pablo Robles-Granda and Jennifer Neville
    On the Challenges of Representing Joint Cumulative Probabilistic GenerativeFunctions of Attributed-Networks
43. Sebastijan Dumancic, Tias Guns, Wannes Meert and Hendrik Blockeel
    Learning Relational Representations with Auto-encoding Logic Programs
44. Timothy van Bremen and Ondrej Kuzelka
    Approximate Weighted First-Order Model Counting: Exploiting Fast Approximate Model Counters and Symmetry
45. Ayush Maheshwari, Hrishikesh Patel, Nandan Rathod, Ritesh Kumar, Ganesh Ramakrishnan and Pushpak Bhattacharyya
    Tale of tails: Rule-Augmented Sequence Labeling for Event Extractionin Low Resource Languages
46. Stefano Teso
    Does Symbolic Knowledge Prevent Adversarial Fooling?

Organizing Committee

• Sebastijan Dumancic (KU Leuven) 
• Angelika Kimmig (Cardiff) 
• David Poole (UBC) 
• Jay Pujara (USC) 

Program Committee

Previous Workshops

Previous StarAI workshops were held in conjunction with AAAI 2010, UAI 2012, AAAI 2013, AAAI 2014, UAI 2015, IJCAI 2016, UAI 2017 and IJCAI 2018 and were among the most popular workshops at the conferences.

• StarAI 2018
• StarAI 2017
• StarAI 2016
• StarAI 2015
• StarAI 2014
• StarAI 2013
• StarAI 2012
• StarAI 2010

Related Venues

• Banff Workshop on New Perspectives for Relational Learning - 2015
• SRL 2012, SRL 2009, SRL 2006, SRL 2004, SRL 2003, and SRL 2000
• NIPS Workshop on Probabilistic Programming - 2008
• Dagstuhl Seminar on Probabilistic, Logical and Relational Learning - 2007, 2005
• Neural-Symbolic Learning and Reasoning