Joint workshop with 2nd Workshop on Natural Language Processing and Automated Reasoning (NLPAR2015)

The 2nd Workshop on Learning and Nonmonotonic Reasoning (LNMR 2015) Natural Language Processing  and Automated Reasoning (NLPAR) will be held jointly with the 2nd Worshop on Natural Language Processing and Automated Reasoning (NLPAR) and will be colocated with the 13th International Conference on Logic Programming and Nonmonotonic Reasoning (LPNMR) in Lexington, Kentucky, USA.

September 2015, Lexington, KY, USA

Program Available here (new!)

Aims and scope

Knowledge representation and reasoning (KR&R) and machine learning are two important fields in artificial intelligence (AI). (Nonmonotonic) logic programming (NMLP) and answer set programming (ASP) provide formal languages for representing and reasoning with commonsense knowledge and realize declarative problem solving in AI. On the other side, inductive logic programming (ILP) realizes inductive machine learning in logic programming, which provides a formal background to inductive learning and the techniques have been applied to the fields of relational learning and data mining. Generally speaking, NMLP and ASP realize nonmonotonic reasoning while lack the ability of (inductive) learning. By contrast, ILP realizes inductive machine learning while most techniques have been developed under the classical monotonic logic. With this background, some researchers attempt to combine techniques in the context of nonmonotonic inductive logic programming (NMILP). Such combination will introduce a learning mechanism to programs and would exploit new applications on the NMLP side, while on the ILP side it will extend the representation language and enable to use existing solvers. Cross-fertilization between learning and nonmonotonic reasoning can also occur in areas including but not limited to:

This workshop follows from its first edition in 2013 in an attempt to provide an open forum for the identification of problems and discussion of possible collaborations among researchers with complementary expertise. To facilitate interactions between researchers in the areas of (machine) learning and nonmonotonic reasoning, we welcome contributions focusing on problems and perspectives concerning both learning and nonmonotonic reasoning.

Keynote

Antonis Kakas on Cognitive Programming

Abstract:

Cognitive Programming is a computing paradigm emerging at the crossroads of Cognitive Psychology and Artificial Intelliegence.
It rests on the synthesis of knowledge from Cognitive Psychology about the structure and activation of human knowledge together with theoretical models and computational methods in Artificial Intelligence in order to provide a natural problem solving environment for ordinary common sense tasks.
Cognitive Programming is strongly driven by the need for computer systems and applications that exhibita smart behaviour for ordinary day to day human tasks. Such applications are generally in the form of Cognitive Assistants that, like a human personal assistant, can help their owner to make decisions that are attuned to the personal preferences of the owner, e.g. to seemingly manage the incoming calls of a user's mobile phone according to the current circumstances of the user and her/his personal preferences, or to focus the search on the Web viaa deeper and personalized understanding of the query posed by the user. The aim is for the programming framework to be naturally simple and highly resilient so that programming can be carried out by the users themselves without the need for any specialized programming skills.

The talk will present the desirable characteristics of Cognitive Programming in contrast with conventional programming frameworks and the main challenges that these requirements pose. It will expose a new perspecitve on the foundational role of logic in programming and discuss the non-monotonic learning tasks that are intrically involved both at the general level of the development of a Cognitive Programnming language and the specific level of the user's development of its own cognitive programs.
These general concepts will be illustrated through a first example of a Cognitive Programming framework, called STAR, that is built by bringing together non-monotonic reasoning, reasoning about actions and belief revision within a logic based argumentation frameworkand adopting a cognitive perspective to logical inference. The talk will show how the STAR system is applied to the problem of automating the cognitive task of text (story) comprehension needed in the developement of cognitive assistants.

Short bio:

Antonis Kakas is a professor at the Department of Computer Science of the University of Cyprus. He obtained a BSc in Mathematics from Imperial College, London in 1980 and then in 1984 a Ph.D. in Theoretical Physics also from Imperial College, London. Since 1989 he is working in Computational Logic and Artificial Intelligence (AI). He is one of the founders of the area of Abductive Logic Programming with his main research interests being in abduction, argumentation and induction. He has worked on the integration of abductive and inductive logic programming with applications in the areas of multi agent systems, reasoning about actions and change and computational bioscience. He has co-edited the book “Abduction and Induction: Essays on their Relation and Integration”. He has co-developed several research software systems for Abductive Reasoning (ACLP & A-system), Argumentation (Gorgias) and Reasoning about Actions and Change (E-RES) which are publicly available on the web. Recently, with others he has proposed, Argumentation Logic, as a logic that offers a new perspective for logic-based AI that facilitates the synthesis of Cognitive Psychology with theory and methods from AI. This has been applied to the problem of Story Comprehension and a new system, called STAR, has been developed for the empirical evaluation of the approach. Currently, he is working on the development of a new framework of Cognitive Programming that aims to offer a programming environment that can be naturally used by human users at large.

Joint Program (new!)

NLPAR

15:40 - 16:00 Break

LNMR

Submissions

We solicit original papers which are not published elsewhere. Papers should be written in English and be formatted according to the Springer Verlag LNCS style, which can be obtained from http://www.springeronline.com. Every paper should not exceed 12 pages including the title page, references and figures. All submissions will be peer-reviewed and all accepted papers must be presented at the workshop.

Papers are to be submitted via the Easychair LNMR15 webpage.

Proceedings

Workshop organizers are considering to publish an on-line proceedings in a formal way. The details will be announced later. Based on the quality of submissions, a special journal issue will also be considered.

Important Dates

Paper registration (new!): June 22 July 6
Submission deadline (new!): June 29 July 10
Notification: August 17
Final version due: September 1
Workshop: 1 or 2 days in September 27-30

Workshop co-Chairs

Alessandra Mileo, INSIGHT Centre for Data Analytics, NUI Galway, Ireland
Alessandra Russo, Dept. of Computing, Imperial College London, UK

Programme Committee