|Molham Aref, RelationalAI, USA — Invited talk sponsored by RelationalAI|
|Relational Artificial Intelligence|
|Abstract: In this talk, I will make the case for a first-principles approach to machine learning over relational databases that exploits recent development in database systems and theory. The input to learning classification and regression models is defined by feature extraction queries over relational databases. We cast the machine learning problem as a database problem by decomposing the learning task into a batch of aggregates over the feature extraction query and by computing this batch over the input database. The performance of this approach benefits tremendously from structural properties of the relational data and of the feature extraction query; such properties may be algebraic (semi-ring), combinatorial (hypertree width), or statistical (sampling). This translates to several orders-of-magnitude speed-up over state-of-the-art systems.
This work is based on collaboration with Hung Q. Ngo (RelationalAI), Mahmoud Abo-Khamis (RelationalAI), Ryan Curtin (RelationalAI), Dan Olteanu (Oxford), Maximilian Schleich (Oxford), Ben Moseley (CMU), and XuanLong Nguyen (Michigan) and other members of the RelationalAI team and faculty network.
Biography: Molham Aref is the Chief Executive Officer of RelationalAI. He has more than 28 years of experience in leading organizations that develop and implement high value machine learning and artificial intelligence solutions across various industries. Prior to RelationalAI he was CEO of LogicBlox and Predictix (now Infor), Optimi (now Ericsson), and co-founder of Brickstream (now FLIR). Molham held senior leadership positions at HNC Software (now FICO) and Retek (now Oracle).
|Michael Gelfond, Texas Tech University, USA — Invited talk sponsored by Potassco Solutions — Joint talk with LPNMR 2019|
|Logic Programming and Non-monotonic Reasoning from 1991 to 2019: a Personal Perspective|
|Abstract: The field of logic programming and nonmonotic reasoning was born in 1991, when a number of researchers working in “the theoretical ends” of logic programming and artificial intelligence gathered in Washington D.C. for the first LPNMR workshop, which was organized by Anil Nerode, Wiktor Marek, and V.S. Subrahmanian. I was privileged to attend this meeting; to closely observe the development of the field over the past 28 years; and to witness many remarkable achievements, which in 1991 I would not have believed to be possible. In this talk I plan to discuss some of these achievements and share a number of personal observations on the field’s history, current state, and possible future directions. Among other things, I will comment on the development of powerful knowledge representation languages, the design and implementation of non-monotonic reasoning systems, and use of these languages and systems in formalizing various types of knowledge and reasoning tasks. The talk is not meant to be a survey of the field, rather it is my personal perspective limited to a small, but important, collection of topics I am most familiar with.
Biography: Dr. Michael Gelfond is a Professor Emeritus in the Computer Science Department at Texas Tech University. He received the M.Sc. from St. Petersburg University, Russia in 1968 and a Ph.D. from the Institute of Mathematics of the Academy of Sciences, St. Petersburg in 1974. His main contributions are in Knowledge Representation and Reasoning and in Declarative Programming. He is best known for his contribution to the development of the Stable Model/Answer Set Semantics of Logic Programming, the Answer Set Programming Paradigm, and Theories of Action and Change. Dr Gelfond is a fellow of the AAAI and the recipient of three Test of Time awards for most influential papers by the International Association of Logic Programming.
|Francesco Ricca, University of Calabria, Italy — Invited tutorial|
|An Extension of Datalog for Modelling and Solving Complex Combinatorial Problems|
|Abstract: Answer set programming (ASP) is a prominent extension of Datalog that allows one for modelling combinatorial problems of high complexity. The success of ASP is due to the combination of two factors: a rich modeling language and the availability of efficient ASP implementations. In this talk we introduce ASP starting from Datalog discussing the role of the main constructs present in the standard ASPCore2 language. Moreover, we overview programming tools and a number of real world applications of ASP.
Biography: Francesco Ricca is currently an Associate Professor at the Department of Mathematics and Computer Science of the University of Calabria, Italy. In the same Department he is Coordinator of the Computer Science Courses Council. He received his Laurea Degree in Computer Science Engineering (2002) and a PhD in Computer Science and Mathematics (2006) from the University of Calabria, Italy, and received the Habilitation for Full Professor in Computer Science (INF/01) in 2017. He is interested in declarative logic-based languages, consistent query answering, and rule-based reasoning on ontologies and in particular on the issues concerning their practical applications: system design and implementation, and development tools. He is co-author of more than 100 (peer-reviewed) publications including international research journals (30+), encyclopedia chapters, conference proceedings, and workshops of national and international importance. He has served in program committees of international conference and workshop, such as IJCAI, AAAI, KR, ICLP, LPNMR and JELIA, and has been reviewer for AIJ, JAIR, TPLP, JLC, etc. He is Area Editor of Association for Logic Programming newsletters, and member of the Executive Board of the Italian Association for Artificial Intelligence.
|Torsten Schaub, University of Potsdam, Germany — Invited talk|
|Dynamic and Temporal Answer Set Programming on Linear Finite Traces (joint work with Pedro Cabalar)|
|Abstract: The logical foundations of Answer Set Programming (ASP) rest upon the logic of Here-and-There (HT), or more precisely its equilibrium models that correspond to stable models in ASP’s semantics. For defining extensions to ASP from firm logical principles, it has thus become good practice to first elaborate upon them in the setting of HT in order to afterwards consider the respective language fragments that are well suited in the context of logic programming.
Inspired by the work on temporal extension of HT over infinite traces, we explore the extension of HT (and ASP) with language constructs from temporal and dynamic logic over linear finite traces. This approach has several advantages. First, it is readily implementable via (incremental) ASP technology. Second, it can be reduced to a normal form close to logic programs. Finally, the finiteness of its models offers a one-to-one correspondence to plans.
We will introduce both extensions, sketch their interrelationships, and their implementation in the temporal extension of the ASP solver CLINGO, dubbed TELINGO and available at github.com/potassco/telingo.
Biography: Torsten Schaub has become a fellow of the European Association for Artificial Intelligence EurAI in 2012. In 2014 he was elected President of the Association of Logic Programming. He served as program (co-)chair of LPNMR’09, ICLP’10, and ECAI’14. The research interests of Torsten Schaub range from the theoretic foundations to the practical implementation of reasoning from incomplete, inconsistent, and evolving information. His current research focus lies on Answer set programming and materializes at potassco.org, the home of the open source project Potassco bundling software for Answer Set Programming developed at the University of Potsdam.