AIJ Prominent and Classic Award Winning Papers


2019 CLASSIC PAPER AWARD was given to.

Between MDPs and semi-MDPs: A framework for temporal abstraction in reinforcement learning

RS Sutton, D Precup, S Singh - Artificial intelligence, 1999 - Elsevier



2019 PROMINENT PAPER AWARD was given to:

The dropout learning algorithm

P Baldi, P Sadowski - Artificial intelligence, 2014 - Elsevier



2018 CLASSIC PAPER AWARD was given to:

Phan Minh Dung
On the Acceptability of Arguments and its Fundamental Role in Nonmonotonic Reasoning, Logic Programming and n-Person Games.
Art. Intell. 77(2): 321-358 (1995)

This is the seminal paper on argumentation theory that laid the foundations for almost all subsequent work in the area. This rich and elegant argumentation framework is developed from a few simple abstract primitives and is used to establish a crisp and meaningful relation between argumentation and theories of non-monotonic reasoning, logic programming, social choice, and cooperative games.

2018 PROMINENT PAPER AWARD was given to:

Martin Gebser, Benjamin Kaufmann, Torsten Schaub:
Conflict-driven answer set solving: From theory to practice.
Artif. Intell. 187: 52-89 (2012)

Answer set programming (ASP) provides a powerful compact language for expressing a number
of combinatorial problems. The paper  introduces a novel  approach for  computing answer sets of
logic programs which is  based on concepts successfully applied in Satisfiability (SAT) checking.
The  approach is implemented in the ASP solver clasp  that has won several contests while
extending  the range of problems that can be modeled and solved effectively as answer set programs.



2017 PROMINENT PAPER AWARD - is shared by two papers this year:

BabelNet: The automatic construction, evaluation and application of a wide-coverage multilingual semantic network. 
Roberto Navigli, Simone Paolo Ponzetto
Artif. Intell. 193: 217-250 (2012) 


YAGO2: A spatially and temporally enhanced knowledge base from Wikipedia. 
Johannes Hoffart, Fabian M. Suchanek, Klaus Berberich, Gerhard Weikum
Artif. Intell. 194: 28-61 (2013).

The prominent paper award for 2017 goes to two papers that have made outstanding outstanding contributions to the automatic construction of large knowledge bases and semantic networks from public domain sources such as Wikipedia and Wordnet. 

BabelNet is a multilingual lexical-semantic network that merges lexicographic information from open-source dictionaries.  Thanks to BabelNet, computational lexical semantic tasks such as disambiguation, information extraction, question answering, and, more in general, text understanding, can be performed for a large number of languages while at the same time preserving the connections across languages, so as to enable the joint analysis and disambiguation of text in multiple languages. 

YAGO2 is  a conveniently searchable, large-scale, highly accurate knowledge base of common facts in machine-processable form. It is invaluable for making sense of internet content and for supporting tasks such as semantic search and text disambiguation, and, in general, for building truly intelligent agents. The YAGO2 paper builds on the earlier YAGO system and focusses on the integration and construction of the spatial and temporal dimension in knowledge bases. 

Both  BabelNet and YAGO2 combine excellent science with a significant engineering effort.
They are not only enabling but also inspiring numerous other developments in artificial intelligence, natural language processing and the semantic web.

The 2017 CLASSIC PAPER AWARD was given to:

Fast Planning Through Planning Graph Analysis. 
Avrim Blum, Merrick L. Furst
Artif. Intell. 90(1-2): 281-300 (1997)

This seminal paper changed the perspective on classical planning algorithms. 

Before the paper appeared, most of the planning approaches used back-chaining methods searching in plan space. Blum and Furst instead proposed to create a particular graph structure in an iterative deepening fashion for constraining
 a backward search from the goal, leading to a dramatic performance increase. Although the specific planning algorithm proposed by the authors did not prevail, the ideas behind the algorithm and the empirical methodology adopted, inspired 
current approaches such as SAT-based planning and heuristic search-based planning methods. The work also demonstrated that it is quite worthwhile to go off the beaten path and take a fresh view on existing algorithmic problems.


 The 2016 PROMINENT PAPER AWARD was given to:

Monte-Carlo tree search and rapid action value estimation in computer Go
Sylvain Gelly and David Silver
Artificial Intelligence 175 (11), Pages 1856-1875, 2011

Go is an ancient Chinese board game that has long been considered one of
the great challenges in AI.

While for a number of years computer programs have managed to beat the world's
leading human players in games like checkers and chess, the high level of intuition
and evaluation required by Go made it tough for AI search methods to crack. This has
changed in recent years, where the last milestone was the recent defeat of the
legendary Go player Lee Se-dol by AlphaGo, a program developed by Google DeepMind.
The previous milestone, however, that enabled this breakthrough, was achieved a decade
ago through the use of Monte-Carlo Tree Search augmented with a number of
enhancements. This led to programs that defeated human professional players and
achieved master (dan) level in 9x9 Go.

That work was reported in 2007 in two papers: one by Rémi Coulom in the
Int. Computer Games Association Journal, the other by Sylvain Gelly and David Silver
at the ICML 2007 conference. This AIJ paper is a follow-up of the latter, covering two of
the enhancements: rapid action value estimation and heuristic initialization. These extensions
led to a program that achieved master level in 19x19 Go for the first time.

The 2016 CLASSIC PAPER AWARD was given to:

Real-time heuristic search
Richard E. Korf
Artificial intelligence 42 (2-3), pp 189-211, 1990

This is the seminal paper in real-time heuristic search over the basic state
model considered in AI, where actions have deterministic effects and
information is complete. While standard heuristic search methods are
aimed at solving problems off-line, real-time search methods are used on-line
for selecting the next action to do after some form of lookahead.

Korf's minimin and real-time A* algorithms take inspiration in the ideas underlying
search in 2-player games, while Learning Real-time A* (LRTA*), where the heuristic
values are updated dynamically during the search, was the first to capture two key
properties: avoidance of loops and convergence to the optimal solution. LRTA* remains
a key reference in the field, where a number of variants have been developed that work
under slightly different assumptions, including the Real-time dynamic programming
algorithm (RTDP), that can be regarded as generalization of LRTA* to Markov Decision Processes.



 The 2015 PROMINENT PAPER AWARD was given to:

Label ranking by learning pairwise preferences
Eyke Hüllermeier , Johannes Fürnkranz , Weiwei Cheng
, Klaus Brinker
Artificial Intelligence, Volume 172, issues  16-17,  November 2008,  pages 1897-1916   

This paper is a key paper in the area of preference learning. 
It studies the problem of label ranking, which is concerned with learning a mapping from instances to rankings over a finite number of labels. 
The authors introduce the Ranking by Pairwise Comparison algorithm (RPC), which first induces a binary preference relation and then uses this relation to derive a ranking. The paper contains appealing theoretical results (that RPC can minimize different loss functions) as well as empirical results (that RPC is competitive in terms of accuracy and superior in terms of efficiency).
The paper shows the elegance and power of a natural and intuitively appealing approach.
It has been influential in the field of preferences and preference learning.

The 2015 CLASSIC PAPER AWARD was given to:

Fusion, Propagation, and Structuring in Belief Networks
Judear Pearl
Artificial Intelligence 29 (3) (1986) 241-288

This is the seminal journal paper that introduced Bayesian networks and the
distributed, linear-time, message-passing algorithm for belief propagation in
singly-connected networks (including trees) . This work along with Pearl's  1988 book,
"Probabilistic Reasoning in Intelligent Systems", sparked what some call the "probabilistic
revolution" in AI. The impact of Bayesian networks and Bayesian networks algorithms  on
AI, Machine Learning, Information Theory, and Cognitive Science has been huge indeed, providing a
representational and computational framework that relates probabilistic reasoning with graphs, graph topology
with complexity bounds, and causal and evidential inference with directional information flow.
By showing  "how to do with probabilities what people say that you can't", the paper introduced
key conceptual notions like the use of graphs for representing  independence relations, and
the use of independence relations for making exact probabilistic inference tractable on tree and
tree-like graphs.  


The 2014 PROMINENT PAPER AWARD was given to the following two papers:

Reasoning about preferences in argumentation frameworks
Sanjay Modgil
173 (9–10), June 2009, Pages 901–934

Argumentation is concerned with attempting to obtain rationally justifiable positions in the presence of conflicting evidence. Originating from the field of philosophy, argumentation research has now become a major topic for AI researchers. One of the key problems in argumentation is to develop a formal model and associated semantics for argumentation that can express the subtleties and nuances of argument and debate. Sanjay Modgil's paper made a major contribution to this problem.

His paper demonstrates how the canonical graph-based models used in abstract argumentation can be enriched to allow such notions as meta-argument, in which arguments can attack attacks. The paper motivates and presents this new model, and explores the relationship of the model to logic programming. Modgil's work represents a key contribution to the argumentation domain, and an outstanding exemplar of work in this area.

Practical solution techniques for first-order MDPs
Scott Sanner and Craig Boutilier,
Artificial Intelligence 173 (5–6), April 2009, Pages 748–788

Decision-theoretic planning problems are naturally represented using probabilistic first-order logic (e.g. PDDL) but are traditionally solved by first 'grounding' the problem. Unfortunately, such a ground representation grows polynomially with the number of domain objects and exponentially in predicate arity. In this seminal paper first-order MDPs are solved without grounding. Although the paper is wide-ranging and could serve as an introduction to this area, it also has the necessary technical depth, providing a clear explanation of solving techniques based on (i) symbolic dynamic programming and (ii) first-order linear programs. Moreover these techniques are implemented and empirically evaluated, showing good results on a range of planning problems. Representing and reasoning with first-order probabilistic theories (often called "lifted inference") is a key research topic in AI; this paper constitutes a major advance to it.

The 2014 CLASSIC PAPER AWARD was given to:

A logic for default reasoning
Ray Reiter 
Artificial Intelligence 13 (1-2), Pages 81-132 (1980) 

This seminal paper introduces and develops a mathematical theory of reasoning about defaults and exceptions that has become to be known as default logic. Reasoning about defaults is about drawing plausible conclusions in the absence of complete knowledge about a world. Default reasoning is a key component of everyday commonsense reasoning, and is essential in many computer systems. 

The central element of default logic is the definition of the extensions to a first-order theory induced by a set of defaults.  Default logic is nonmonotonic, in the sense that conclusions justified by defaults may need to be retracted when new axioms are added.  

Reiter's approach to default reasoning has been immensely influential. Not only it has made a significant impact on the field of knowledge representation, including its application  to the frame problem and to other difficult issues in the theory of commonsense reasoning, but it has also made a significant impact on logic programming and underlies much current work on default reasoning including answer set programming. 

Overall, this article is one of the cornerstone publications of the knowledge representation research domain, and indeed of AI in general. The award committee is  privileged to have the opportunity to recommend unanimously this paper as the recipient of the 2014 AIJ Classic Paper Award.



The 2013 PROMINENT PAPER AWARD was given to:

Combining answer set programming with description logics for the Semantic Web
Thomas Eiter, Giovambattista Ianni, Thomas Lukasiewicz, Roman Schindlauer, Hans Tompits 
Artificial Intelligence 172(12-13): 1495-1539 (2008)

This paper proposes dl-programs, a formalism that integrates description logics with rule-based logic programs under answer set semantics. It provides not only detailed theoretical analyses of these programs in terms of their expressive power and computational complexities, but also an implementation that illustrates the usefulness of the proposed formalism in the semantic web. This work highlights many difficult issues in the problem of adding rules and default rules into description logics, and has been very influential in subsequent work in this area.


The 2013 CLASSIC PAPER AWARD was given to the following two papers:

Richard Fikes, Nils J. Nilsson
STRIPS: A New Approach to the Application of Theorem Proving to Problem Solving.
Artificial Intelligence 2(3/4): 189-208 (1971)

This paper lays the foundations and initial algorithms for what has become to be known as classical planning in AI  where an agent has to perform deterministic actions for transforming a given initial initial state into a goal state from a declarative and compact representation of the actions. For this, the paper combines ideas from logic and problem solving in the formulation of a domain-independent problem solver where the states are characterized by first-order logical formulas, and operators  are characterized by  three sets of formulas --  the precondition, add, and delete lists. The  representation provides a practical solution to the frame problem, which  with some variations,  is still in use in current classical and non-classical planners alike. The basic STRIPS planning algorithm provides in turn the basis for linear  and non-linear planning algorithms, and for the view of domain-independent classical planning as a path-finding problem in the graph of states.


Alan K. Mackworth
Consistency in Networks of Relations
Artificial Intelligence 8(1):99-118 (1977)

This seminal paper in the field of AI devoted to solving constraint satisfaction problems (CSPs), contains three foundational contributions. First, the paper contributes a fundamental insight for improving the performance of backtracking algorithms on CSPs by identifying that local inconsistencies can lead to much thrashing or unproductive search. Second, the paper presents clear definitions of conditions that characterize the level of local consistency of a CSP, notably including the concept of arc consistency, and precise algorithms for enforcing these levels of local consistency by removing inconsistencies. Such algorithms have come to be known as constraint propagation algorithms. Third, the paper advocates the use of constraint propagation at each node in the search tree, a technique that is now the foundation of all open source and commercial constraint programming systems.  The paper has been immensely influential in establishing, and guiding the research agenda of, the field of constraint programming.


The 2012 PROMINENT PAPER AWARD was given to:

Learning and inferring transportation routines
Lin Liao, Donald J. Patterson, Dieter Fox, and Henry Kautz
Vol 171 (5–6), April 2007, Pages 311–331

This paper introduces a hierarchical Markov model that can learn and  infer a user’s daily movements through an urban community, and applies it in an application that helps cognitively-impaired people use public transportation safely. The paper takes a realistic and important  problem, and solves it by developing technically sophisticated,  state-of-the-art AI techniques, that have applicability well beyond the domain described in the paper. This work has had a significant impact on the  area of modeling and learning with dynamic Bayesian networks, both in  and outside of AI. As such, the award committee unanimously believes the  paper is a worthy winner of the inaugural AIJ Prominent Paper Award.


There was not yet any CLASSIC PAPER AWARD in 2012.