Showing papers in "Artificial Intelligence in 2014"

TL;DR: In this paper, the authors describe extensions and improvements of existing models, new families of models, and a much more thorough treatment of algorithm parameters as model inputs, and comprehensively describe new and existing features for predicting algorithm runtime for propositional satisfiability (SAT), travelling salesperson (TSP), and mixed integer programming (MIP) problems.

TL;DR: A mathematical analysis of some of the static and dynamic properties of dropout is provided using Bernoulli gating variables, general enough to accommodate dropout on units or connections, and with variable rates, which allows a complete analysis of the ensemble averaging properties ofdropout in linear networks.

TL;DR: Empirical evaluation demonstrates that general appliance models can be constructed using data from only a small number of appliances, and furthermore that 28–99% of the remaining behaviour which is specific to a single household can be learned using only aggregate data from existing smart meters.

TL;DR: This work proposes a novel machine learning framework called "Online Transfer Learning" (OTL), which aims to attack an online learning task on a target domain by transferring knowledge from some source domain, and investigates two different settings of OTL: (i) OTL on homogeneous domains of common feature space, and (ii) O TL across heterogeneous domainsof different feature spaces.

TL;DR: A novel multi-armed bandit (MAB) model, the bounded MAB is introduced, and an algorithm to solve it efficiently, called bounded e-first, is developed, which outperforms existing crowdsourcing methods by up to 300%, while achieving up to 95% of a hypothetical optimum with full information.

TL;DR: This paper advocates a change in focus to actors as the primary topic of investigation, and discusses open problems and research directions toward that objective in knowledge representations, model acquisition and verification, synthesis and refinement, monitoring, goal reasoning, and integration.

TL;DR: A competence-based concept detection method that requires no prior knowledge of case distribution and provides statistical guarantees on the reliability of the changes detected, as well as meaningful descriptions and quantification of these changes.

TL;DR: In this paper, a generic approach for reasoning over argumentation frameworks is proposed based on the concept of complexity-sensitivity, which allows instantiations of the generic framework via harnessing in an iterative way current sophisticated Boolean satisfiability solver technology for solving the considered argumentation reasoning problems.

TL;DR: It is suggested in the context of question-answering that what matters when it comes to the science of AI is not a good semblance of intelligent behaviour at all, but the behaviour itself, what it depends on, and how it can be achieved.

TL;DR: It is shown that positive existential and nonrecursive datalog rewritings, which do not use extra non-logical symbols, suffer an exponential blowup in the worst case, while first-order rewrites can grow superpolynomially unless NP ⊆ P/poly.

TL;DR: This work formally introduces a novel weighted synergy graph model to capture new interactions among agents and shows that the weighted synergygraph outperforms a competing algorithm, thus illustrating the efficacy of the model and algorithms.

TL;DR: A novel algorithm framework, called TRAMP, is explored, to learn action models with limited training data in a target domain, via transferring as much of the available information from other domains as possible to help the learning task, assuming action models in source domains can be transferred to the target domain.

TL;DR: This empirical study considers various scenarios including random, realistic, and structured problems and reveals the advantage of the use of the proposed heuristics in the anytime framework over state-of-the-art local search algorithms.

TL;DR: In this article, the complexity of manipulation algorithms for elections over nearly single-peaked electorates is studied, and it is shown that even one maverick jumps the manipulation complexity up to NP-hardness.

TL;DR: The CPD problem is introduced and some of the main approaches that have been used by structural biologists to solve it, with an emphasis on the exact method embodied in the dead-end elimination/A? algorithm (DEE/A?).

TL;DR: In this paper, the authors consider the problem of inconsistency management in multi-context systems, in which decentralized and heterogeneous system parts interact via (possibly nonmonotonic) bridge rules for information exchange.

TL;DR: This work examines analogous notions to the problem of enumerating preferred extensions and presents algorithms that decide subjective, respectively objective, acceptance in a widely studied development of Dung@?s model: namely value-based argumentation frameworks (vafs).

TL;DR: This paper provides a comprehensive set of theoretical results putting the intuitive notion of Input/Output behavior of argumentation frameworks on a solid formal ground and develops an extensive analysis of all the concepts listed above, covering seven well-known literature semantics and taking into account various, more or less constrained, ways of partitions an argumentation framework.

TL;DR: A new learning algorithm, called HTNLearn, is proposed to help acquire HTN methods and action models simultaneously from partially observed plan traces, and the experimental results show that the algorithm HTN learn is both effective and efficient.

TL;DR: KLog as discussed by the authors is a kernel-based approach to statistical relational learning, which allows users to specify learning problems declaratively and allows access by the kernel to the rich representation mediated by a technique called graphicalization.

TL;DR: This paper formally stating the precise relation between several constraint answer set programming languages - AC, clingcon, ezcsp - as well as the underlying systems is addressed.

TL;DR: Novel results concerning the incremental computation of the stable and ideal argumentation semantics are provided, by extending the division-based method presented in [1] and correcting a flawed claim given there.

TL;DR: Experimental results demonstrate that teaching guidance substantially improves human teaching in three different problem domains, providing promising evidence that human intelligence and flexibility can be leveraged to achieve better sample efficiency when input data to a learning system comes from a human teacher.

TL;DR: This work embraces pure-strategy Nash equilibrium, an important solution concept in non-cooperative game theory, to formally define the stable outcomes of a network influence game and to predict potential outcomes without explicitly considering intricate dynamics.

TL;DR: Using Possible Conflicts, a structural model decomposition method from the Artificial Intelligence model-based diagnosis (DX) community, a distributed diagnoser design algorithm is developed to build local event-based diagnosers that are constructed based on global diagnosability analysis of the system.

TL;DR: It is proved that it is NP-hard for a coalition of two manipulators to compute a manipulation of unweighted Borda voting, and several new heuristic methods are proposed that attempt to minimise the number of manipulators.

TL;DR: In this paper, it was shown that actors occupying the same social position based on regular equivalence will satisfy the same set of modal formulas, and analogous results for regular similarity and generalized regular equivalences based on many-valued modal logics.

TL;DR: This article develops a quadratic representation of EL TBoxes as regular tree grammars and shows that, if a finite uniform EL interpolant exists, then there exists one that is at most triple exponential in the size of the original TBox, and that, in the worst case, no smaller interpolants exist, thereby establishing tight worst-case bounds on their size.

TL;DR: This paper enhances theFLP answer set semantics with a level mapping formalism such that every answer set I can be built by fixpoint iteration of a one-step provability operator (more precisely, an extended van Emden-Kowalski operator for the FLP reduct fΠI) and shows that the level mapping does not increase the worst-case complexity of FLP answer sets.

TL;DR: The results suggest that when compared with previous anytime algorithms, APTS/ANA?