Saturday July 9th

Room 1 (GT/MAS/Agents)
8:45-10:30
T1 - Bitcoin: a Basic Tutorial on Decentralized Digital Currencies
Aviv Zohar, The Hebrew University of Jerusalem
Bitcoin is a disruptive new protocol for digital money that has the potential to revolutionize money transmission. This tutorial will cover the inner workings of the protocol, and introduce participants to important research questions related to crypto-currencies.

11:00-12:45
T6- The Internet of Things and Multiagent Systems
Munindar P. Singh, North Carolina State University; Amit K. Chopra, Lancaster University
This tutorial introduces the Internet of Things (IoT), a rapidly expanding technology area. It describes how ideas from artificial intelligence-specifically, multiagent systems-can support the IoT as well as additional research advances needed in the relevant areas to help realize the IoT.

1:45-3:30
T7 - Organ Exchanges: A Success Story of AI in Healthcare
John Dickerson, Carnegie Mellon University; Tuomas Sandholm, Carnegie Mellon University
This tutorial overviews a fielded example of technology at the intersection of AI and economics: organ exchange, a method by which patients in need of a organ can swap willing but incompatible donors. It covers past and current research in organ exchange; throughout, it also gives a higher-level overview of the steps taken to translate a purely academic idea into a large fielded healthcare system.

4:00-5:45
T7 - Organ Exchanges: A Success Story of AI in Healthcare
John Dickerson, Carnegie Mellon University; Tuomas Sandholm, Carnegie Mellon University


Room 2 (ML)
8:45-10:30
T2 - Information-theoretic Ideas in Machine Learning
Greg Ver Steeg, Information Sciences Institute; Aram Galstyan, Information Sciences Institute
While information theory was developed to solve engineering challenges in communication, the fundamental principles seem to have a much broader reach. This tutorial will review some of the challenges and successes in applying information-theoretic concepts in the context of machine learning.

11:00-12:45
T2 - Information-theoretic Ideas in Machine Learning
Greg Ver Steeg, Information Sciences Institute; Aram Galstyan, Information Sciences Institute

1:45-3:30
T8 - Statistical Relational AI: Logic, Probability, Computation
Luc De Raedt, KU Leuven; David Poole, University of British Columbia; Kristian Kersting, University of Dortmund; Sriraam Natarajan, Indiana University
This tutorial will provide a gentle introduction into the foundations of statistical relational artificial intelligence, and will realize this by introducing the foundations of logic, of probability, of learning, and their respective combinations.

4:00-5:45
T8 - Statistical Relational AI: Logic, Probability, Computation
Luc De Raedt, David Poole, Kristian Kersting, Sriraam Natarajan

Room 3 (Logic/KR)
8:45-10:30
T3 - Belief functions for the working scientist
Fabio Cuzzolin, Oxford Brookes University
All fields of Artificial Intelligence and applied science are subject to various degrees of uncertainty, caused by missing or scarce data: random sets and their subjective incarnations called "belief functions" naturally arise when one tries to formalize lack of data in a coherent way. This tutorial introduces to a wider AI audience the basic principles of the theory of belief functions, describes practical tools based on them and compares their performance against more classical approaches to uncertainty on cutting edge real-world problems.

11:00-12:45
T3 - Belief functions for the working scientist
Fabio Cuzzolin, Oxford Brookes University

1:45-3:30
T9 - Reasoning with Description Logics
Ivan Varzinczak, Universite d'Artois
This tutorial provides an introduction to Description Logics, a family of logic-based knowledge representation formalisms with compelling computational properties and a variety of applications in modern AI.

4:00-5:45
T9 - Reasoning with Description Logics
Ivan Varzinczak, Universite d'Artois


Room 4 (Robotics/Planning)
8:45-10:30
T22 - Optimality in Robot Motion and Action
Jean-Paul Laumond, LAAS-CNRS; Nicolas Mansard, LAAS-CNRS
For robots and living beings, the link between actions expressed in the physical space and motions originated in the motor space, turns to geometry in general and, in particular, to linear algebra. The tutorial surveys more than 40 years of research in both mobile and humanoid robotics tending to express robot actions as motions to be optimized.

11:00-12:45
T22 - Optimality in Robot Motion and Action
Jean-Paul Laumond, LAAS-CNRS; Nicolas Mansard, LAAS-CNRS

1:45-3:30
T10 - How Computers Read the Web
Estevam R. Hruschka Jr., Federal University of Sao Carlos
This tutorial explores some of the most relevant Machine Reading approaches intended to build computer systems designed to Read the Web.

4:00-5:45
T10 - How Computers Read the Web
Estevam R. Hruschka Jr., Federal University of Sao Carlos



Room 5 (Misc)
8:45-10:30
T5 - Deep Learning and Continuous Representations for NLP
Scott Wen-tau Yih, Microsoft Research; Xiaodong He, Microsoft Research; Jianfeng Gao, Microsoft Research
This tutorial provides an extensive overview on recently developed techniques in deep learning and continuous-space representations for natural language problems. Approaches to important real-world natural language applications, including machine translation, semantic representation modeling, question answering and semantic parsing will be the main focus.

11:00-12:45
T5 - Deep Learning and Continuous Representations for NLP
Scott Wen-tau Yih, Microsoft Research; Xiaodong He, Microsoft Research; Jianfeng Gao, Microsoft Research

1:45-3:30
T11 - Multidimensional Text Clustering for Hierarchical Topic Detection
Nevin Zhang, The Hong Kong University of Science and Technology; Leonard K. M. Poon, The Hong Kong Institute of Education
Text clustering is generally considered unsuitable for topic detection because it associates each document with only one “topic” (i.e., document cluster). Recent advances in model-based multidimensional clustering have overcome the difficulty, and have given rise to a novel approach to hierarchical topic detection that outperformed the LDA approach in empirical studies.

4:00-5:45
T12 - Affect Detection from Spoken and Written Text –Computational Models for Affect and Sentiment Analysis
Maite Taboada, Simon Fraser University; Alexandra Balahur, European Commission JRC; Bjorn Schuller, Imperial College London
Automatic affect detection and classification from text is a complex task in Artificial Intelligence, leveraging the use of methods and resources from various fields: Computational Linguistics, Data Mining, Linguistics, but also on theories from Psychology, Cognitive Science, and Social Psychology. The present tutorial gives a broad overview of the issues tackled so far in the field, the methods employed and resources created, as well as the challenges identified and possible avenues for further investigation.

Sunday July 10th

Room 1 (GT/MAS/Agents)
8:45-10:30
T13 - Eliciting high quality information
Boi Faltings, EPFL; Goran Radanovic, EPFL
AI systems often depend on information provided by other agents, whether it is for learning or decision-making. However, in most cases they have no way of verifying that this information is correct and relevant. In this tutorial, we will present game-theoretic techniques that allow to verify and incentivize accurate information by exploiting the multi-agent nature of information elicitation.

11:00-12:45
T19 - Argument and Cognition
Antonis Kakas, Open University of Cyprus; Loizos Michael, Open University of Cyprus
Building machines that are aware of, and properly accommodate for, the cognitive capabilities and limitations of their human users is emerging as a key characteristic of cognitive systems. This tutorial will discuss how the synthesis of work in Formal Argumentation from AI and work in Narrative Text Comprehension from Cognitive Psychology can offer a scientifically sound and pragmatic basis for building human-aware Cognitive AI systems for common everyday tasks.

1:45-3:30
T18 - Generating Synthetic Populations for Social Modeling
Samarth Swarup, Virginia Tech; Madhav V. Marathe, Virginia Tech
This tutorial will introduce an AI audience to methodologies for generating realistic, data-driven, large-scale synthetic populations by integrating data from multiple sources. This methodology is of interest to many topics of emerging importance for AI, including computational sustainability and resilience, urban computing, disaster response, and more.

4:00-5:45
T24 - If Turing had lived longer: how might he have investigated what AI and Philosophy can learn from evolved information processing systems?
Aaron Sloman, Birmingham University
An interactive discussion on what Alan Turing might have studied if he had lived several more decades after publishing his 1952 paper on Morphogenesis. The Meta-Morphogenesis project attempts to identify forms and mechanisms of information processing produced by evolution since our planet formed, including forms of computation possibly related to deficiencies in current AI.

Room 2 (ML)
8:45-10:30
T14 - Lifted Probabilistic Inference in Relational Models
Guy Van den Broeck, UCLA; Dan Suciu, University of Washington, Seattle
This tutorial explains the core ideas behind lifted probabilistic inference in statistical relational learning and probabilistic databases. Both fields deal with relational representations of uncertainty, for which efficient inference is an enormous challenge, yet have also developed efficient algorithms for tasks thought to be highly intractable.

11:00-12:45
T14 - Lifted Probabilistic Inference in Relational Models
Guy Van den Broeck, UCLA; Dan Suciu, University of Washington, Seattle


1:45-3:30
T20 - New Advances in Combinatorial Optimization for Graphical Models
Rina Dechter, University of California, Irvine; Radu Marinescu, IBM Research; Alex Ihler, University of California, Irvine; Lars Otten, Google
This tutorial will present state-of-the-art algorithms for solving combinatorial optimization tasks in different graphical models (Bayesian networks, Markov networks, Constraint networks) and demonstrate their applicability to problems in scheduling, design and diagnosis, bioinformatics tasks, and web-based applications.

4:00-5:45
T25 - Scalable learning of graphical models (CANCELLED)
Francois Petitjean, Monash University; Geoffrey Webb, Monash University
To scale graphical modeling techniques to the size and dimensionality of most modern data stores, data science researchers and practitioners now have to meld the most recent advances in numerous specialized fields including graph theory, statistics, pattern mining and graphical modeling. This tutorial will present the core building blocks that are necessary to build and use scalable graphical modeling technologies on large and high-dimensional data.


Room 3 (Logic/KR)
8:45-10:30
T15 - Rulelog: Rule-based Knowledge Representation and Reasoning
Benjamin Grosof, Coherent Knowledge; Michael Kifer, Stony Brook University; Paul Fodor, Stony Brook University
Rulelog is a major basic research advance in fully semantic knowledge representation and reasoning, that has a wide variety of applications and has capable efficient implementations leveraging methods from logic programming and relational/graph databases. It features: highly expressive higher-order formulas, defeasibility, probabilistic uncertainty, bounded rationality, and other strong meta; polynomial-time computational tractability; close combination with natural language processing; and strong complementarity with machine learning as well as other semantic technologies.

11:00-12:45
T15 - Rulelog: Rule-based Knowledge Representation and Reasoning
Benjamin Grosof, Coherent Knowledge; Michael Kifer, Stony Brook University; Paul Fodor, Stony Brook University


1:45-3:30
T21 - AI for Smarter Cities
Pascal Hitzler, Wright State University; Freddy Lecue, IBM Research; Raghava Mutharaju, Wright State University; Jeff Pan, University of Aberdeen; Jiewen Wu, IBM Research
Cities take more and more advantages of information and communications technology (ICT) to better manage their resources and improve the quality of life of its citizens. ICT spans many departments of the cities, from transportation, water, energy to building management and social-care services. AI techniques are getting more and more attraction from cities to represent and organize information, maintain sustainable networks, predict incidents, optimize distribution, diagnose faults, plan routes and organize their infrastructure. Managing traffic efficiently, among many other domains in cities, is one of the key issues in large cities. In this tutorial we describe the domains of applications which could benefit from AI techniques, along with introducing the necessary background knowledge. Then we focus on trafficc applications, which make use of recent AI research in knowledge representation, logic programming, machine learning, planning, reasoning and optimization. Speciffically we go through the next version of scalable AI driven traffic related application where (1) data from a variety of sources is collected, (2) knowledge about traffic, vehicles, citizens, events is represented and (ii) deductive and inductive reasoning is combined for diagnosing and predicting road traffic congestion. Based on these principles, a real-time, publicly available AI system named STAR-CITY was developed. We discuss the results of deploying STAR-CITY, and its related AI technologies in cities such as Dublin, Bologna, Miami, Rio and the lessons learned. The final part of the tutorial aims at discussing future AI opportunities including scalability issues for large cities.

4:00-5:45
T21 - AI for Smarter Cities
Pascal Hitzler, Wright State University; Freddy Lecue, IBM Research; Raghava Mutharaju, Wright State University; Jeff Pan, University of Aberdeen; Jiewen Wu, IBM Research

Room 4 (Robotics/Planning)
8:45-10:30
T16 - Symbolic and structural models for image understanding
Jamal Atif, LAMSADE, Univ. Paris Dauphine; Isabelle Bloch, LTCI, CNRS, Telecom ParisTech; Celine Hudelot, Centrale-Supelec
In this tutorial, at the cross-road of artificial intelligence and visual information understanding, symbolic and structural models for high level image and scene understanding will be presented, including knowledge representation and reasoning methods. Examples in medical and satellite imaging will illustrate these methods.

11:00-12:45
T16 - Symbolic and structural models for image understanding
Jamal Atif, LAMSADE, Univ. Paris Dauphine; Isabelle Bloch, LTCI, CNRS, Telecom ParisTech; Celine Hudelot, Centrale-Supelec


1:45-3:30
T4 - Spatio-Temporal Stream Reasoning for Safe Autonomous Systems
Fredrik Heintz, Linkoping University
Stream reasoning is the incremental reasoning over streams of incrementally available information. This tutorial gives an overview of the state-of-the-art in efficient and scalable logic-based approaches to spatio-temporal stream reasoning with incomplete information and applications such as execution monitoring of autonomous systems.

4:00-5:45
T4 - Spatio-Temporal Stream Reasoning for Safe Autonomous Systems
Fredrik Heintz, Linkoping University


Room 5 (Misc)
8:45-10:30
T17 - Programming by Optimization: A Practical Paradigm for Computer-Aided Algorithm Design
Holger Hoos, Frank Hutter, Kevin Leyton-Brown
Programming by Optimization (PbO) is a general approach for developing high-performance solvers for challenging computational problems that leverages machine learning and optimization methods to customize a rich and flexible algorithm design space for specific use contexts. In this tutorial, we describe the PbO paradigm and the automated algorithm configuration and selection techniques that enable it, demonstrate how it has yielded dramatic performance gains on a range of AI problems, and show how PbO is supported by a set of readily available tools.

11:00-12:45
T17 - Programming by Optimization: A Practical Paradigm for Computer-Aided Algorithm Design
Holger Hoos, Frank Hutter, Kevin Leyton-Brown


1:45-3:30
T23 - Scalable Big Data Programming Models for AI
John Yen, Penn State University
The tutorial will introduce Spark, a scalable programming model for Big Data AI research and application developments. Through hands-on labs, participants will learn how to harvest AI opportunities in massive datasets by leveraging Spark and related software stacks such as Hadoop Distributed File Systems (HDFS), Yarn, and the Machine Learning Library of Spark. The rapid explosion of data (in terms of volume, velocity, variety) opens up massive opportunities for conducting research and developing applications regarding artificial intelligence. Hence, AI solutions of this big data era need to be scalable for handling massive data. Spark, part of Berkeley Data Analytic Stack (BDAS) developed by the AMPLab at UC Berkeley, offers a highly flexible critical open source that has been shown to be 10 to 100 times faster than MapReduce for iterative algorithms like those used in most AI software. Spark is implemented on Scala, a novel programming language that integrates object-oriented programming, functional programming, and procedural programming in an integrated way. In this tutorial, I will introduce programming concepts and models of Scala and Spark that are designed for scalable data-intensive processing. These features will be demonstrated using hands on labs regarding the analytics of social media data and scholarly data. The tutorial will also introduce related software stacks for Big Data computing such as Hadoop Distributed File Systems, Yarn, Spark Streaming (for processing streaming data), Spark MLlib (for machine learning models), and Spark GraphX (for processing massive heterogeneous networks). Together, they not only enable the development AI innovations by harvesting massive datasets, but also facilitate their deployment as robust and fault tolerant AI applications.

4:00-5:45
T23 - Scalable Big Data Programming Models for AI
John Yen, Penn State University

Monday July 11th

Room 1 (GT/MAS/Agents)
8:45-10:30
T26 - Argument Mining
Serena Villata, CNRS; Katarzyna Budzynska, Polish Academy of Sciences
Nowadays, on the one hand, text analysis is a promising approach to identify and extract arguments from text, receiving attention from the natural language processing community (e.g., argument mining of legal documents, on-line debates, newspaper and scientific articles); on the other hand, computational models of argumentation have made substantial progress in providing abstract and structured formal models to represent and reason over argumentation structures. This tutorial focusses on the interaction between Computational Linguistics and Argumentation Theory with the goal to discuss the techniques and frameworks that have been proposed to analyze, aggregate, synthesize, structure, summarize, and reason about arguments in texts.

11:00-12:45
T26 - Argument Mining
Serena Villata, CNRS; Katarzyna Budzynska, Polish Academy of Sciences


1:45-3:30
T32 - The Soar Cognitive Architecture
Nate Derbinsky, Wentworth Institute of Technology; John Laird, University of Michigan
This tutorial provides a whirlwind tour of Soar, including its historical background and context within the cognitive architecture research community. The focus will be on how and why Soar works, including syntax/structure for agents and Soar-enabled applications, as well as hands-on experience with the new components that have been developed over the past few years: reinforcement learning, semantic/episodic long-term memory, and mental imagery.

4:00-5:45
T32 - The Soar Cognitive Architecture
Nate Derbinsky, Wentworth Institute of Technology; John Laird, University of Michigan


Room 2 (ML)
8:45-10:30
T27 - Pattern Recognition on Random Graphs
Li Chen, Intel Corp
Pattern recognition methodologies developed for random graph models have become increasingly important, as graphs are enjoying much popularity to represent certain datasets. This tutorial will cover vertex classification, clustering, nomination, seeded graph matching, scalable seeded graph matching and joint vertex classification and their applications in social networks, neuroscience and online advertising problems.

11:00-12:45
T31 - Low-Rank and Sparse Modeling for Data Analytics
Sheng Li, Northeastern University; Yun Fu, Northeastern University
Low-rank and sparse modeling are emerging mathematical tools dealing with uncertainties of real-world data. This tutorial will explain the research problems, existing techniques, and future challenges of low-rank and sparse modeling, and will review a wide range of applications such as clustering, semi-supervised classification, multi-view learning, and transfer learning.

1:45-3:30
T33 - Integrating Learning and Search for Structured Prediction
Alan Fern, Oregon State University; Liang Huang, Oregon State University; Jana Doppa, Washington State University
Structured Prediction (SP) is an exciting field with a number of potential applications including natural language processing, computer vision, planning, and bioinformatics. We will survey the existing approaches for SP that integrate machine learning and search: 1) Overview of traditional cost function learning approaches (e.g., CRFs and SSVMs); 2) Control knowledge learning framework and instantiations including greedy methods (classifier-based SP and Easy-first Framework) and beam search methods (structured perceptron with inexact search); and 3) HC-Search framework that unifies both cost function learning and control knowledge learning frameworks, and its instantiations.

4:00-5:45
T33 - Integrating Learning and Search for Structured Prediction
Alan Fern, Oregon State University; Liang Huang, Oregon State University; Jana Doppa, Washington State University



Room 3 (Logic/KR)
8:45-10:30
T28 - Methodologies for Ontology Based Data Access Applications
Giuseppe De Giacomo, Sapienza Univ Roma; Domenico Lembo, Sapienza Univ Roma; Antonella Poggi, Sapienza Univ Roma; Valerio Santarelli, Sapienza Univ Roma; Domenico Fabio Savo, Sapienza Univ Roma
The tutorial illustrates methodologies for developing ontology-based data access (OBDA) applications, which aim at coupling conceptual views of information, expressed as Description Logic ontologies, with actual and possibly pre-existing data stores. In the tutorial we will present the basics of OBDA, introduce a graphical model for quick development of OWL 2 ontologies, survey typical mechanisms to link ontologies with data, and discuss some special reusable patterns for modeling recurrent representation needs. We will conduct an hands-on-session in which participants will develop (small) OBDA applications by applying the methodologies learned and by exploiting state-of-the-art OBDA tools.

11:00-12:45
T28 - Methodologies for Ontology Based Data Access Applications
Giuseppe De Giacomo, Domenico Lembo, Antonella Poggi, Valerio Santarelli, Domenico Fabio Savo

1:45-3:30
T34 - Solving (Problems with) Quantified Boolean Formulas: Recent Trends and Challenges
Florian Lonsing, TU Wien
This tutorial presents quantified Boolean formulas (QBFs) from both a practical and theoretical perspective. It introduces and reviews the state of the art in QBF research and aims at providing non-specialists with the necessary background to apply QBF-based techniques in practice.

4:00-5:45
T37 - Scalable Probabilistic Logics
William Yang Wang, Carnegie Mellon University; William W Cohen, Carnegie Mellon University
Although many information-management tasks (including classification, retrieval, information extraction, and information integration) can be formulated as inference in an first-order logic, most probabilistic first-order logics are not efficient enough to be used for large-scale versions of these tasks. In this tutorial, we provide a gentle introduction to probabilistic logics, and then describe recent advances in designing scalable probabilistic logics, with a special focus on a newly-developed scalable probabilistic logic called ProPPR.

Room 4 (Robotics/Planning)
8:45-10:30
T29 - Introduction to Planning Models and Methods
Hector Geffner, Universitat Pompeu Fabra; Blai Bonet, Universidad Simon Bolivar
Planning is the model-based approach to autonomous behavior where the agent behavior is derived automatically from a model. In the tutorial, we will look at the variety of models used in AI planning and the methods that have been developed for solving them. The goal is to provide a modern and coherent view of planning that is comprehensive but not shallow.

11:00-12:45
T29 - Introduction to Planning Models and Methods
Hector Geffner, Universitat Pompeu Fabra; Blai Bonet, Universidad Simon Bolivar


1:45-3:30
T35 - Deliberative Planning and Acting
Malik Ghallab, LAAS-CNRS; Dana Nau, University of Maryland; Paolo Traverso, FBK ICT IRST
For an agent to interact effectively with a complex environment, deliberation is required during both planning (choosing which actions to perform in which circumstances) and acting (deciding how to perform those actions in the current environment and context). We will provide an up to-date survey of the relevant techniques, including hierarchical, temporal, and nondeterministic methods.

4:00-5:45
T35 - Deliberative Planning and Acting
Malik Ghallab, LAAS-CNRS; Dana Nau, University of Maryland; Paolo Traverso, FBK ICT IRST



Room 5 (Misc)
8:45-10:30
T30 - Coreference Resolution: Recent Successes and Future Directions
Vincent Ng, University of Texas at Dallas
Coreference resolution is one of the most challenging tasks in natural language understanding whose solution requires the acquisition of and reasoning with a variety of linguistic knowledge sources. This tutorial will discuss (1) recent successes and challenges in coreference research and (2) the relevance of coreference resolution to the Winograd Schema Challenge, which has recently received a lot of attention in the AI community.

11:00-12:45
T30 - Coreference Resolution: Recent Successes and Future Directions
Vincent Ng, University of Texas at Dallas


1:45-3:30
T36 - Research Challenges in Computational Sustainability
Stefano Ermon, Stanford; Bistra Dilkin, Georgia Tech
Computational sustainability is a new interdisciplinary research field focused on developing computational models, methods and tools to help solve some of the key sustainability challenges of our times. This tutorial will provide an overview of the field of computational sustainability, introducing a number of examples, research questions, open problems, and computational techniques.

4:00-5:45
T38 - Towards a Unified Framework for Transfer Learning: Exploiting Correlations and Symmetries
Sridhar Mahadevan, University of Massachusetts at Amherst
This tutorial provides a unified framework that reviews recent work on transfer learning, centered around two mathematical themes of maximizing correlations and exploiting symmetries. A range of real-world examples, from image recognition, cross-lingual information retrieval, natural language processing, and reinforcement learning will be used to illustrate the techniques.