Activity recognition is a key problem in multi-sensor systems. In this work, we introduce Computational Assumption-based Argumentation, an argumentation approach that seamlessly combines sensor data processing with high-level inference. Our method gives classification results comparable to machine learning based approaches with reduced training time while also giving explanations.
Ilya Otpuschennikov, Alexander Semenov, Irina Gribanova, Oleg Zaikin, Stepan Kochemazov
1594 - 1595
In this paper we propose the technology for constructing propositional encodings of discrete functions. It is aimed at solving inversion problems of considered functions using state-of-the-art SAT solvers. We implemented this technology in the form of the software system called TRANSALG, and used it to construct SAT encodings for a number of cryptanalysis problems. By applying SAT solvers to these encodings we managed to invert several cryptographic functions. In particular, we used the SAT encodings produced by TRANSALG to construct the family of two-block MD5 collisions in which the first 10 bytes are zeros. In addition to that we used TRANSALG encoding for the widely known A5/1 keystream generator to solve several dozen of its cryptanalysis instances in a distributed computing environment. Also in the present paper we compare the functionality of TRANSALG with that of similar software systems.
Quan Yu, Hai Wan, Jiangtao Xu, Freddy Lécué, Liang Chang
1596 - 1597
Explanatory diagnosis of an ontology stream aims to explain the changes hidden in the ontology stream by a sequence of actions. In this paper, we present a framework for explanatory diagnosis of an ontology stream, which allows the actions to be uncertain. In order to capture the semantics of actions, we introduce a new update operator and effect-guided bold-repair. By combining these operators with a query mechanism of description logics supporting inconsistency-tolerant semantics, we present a formal definition for the explanatory diagnosis problem of ontology streams.
Thomas Bridi, Michele Lombardi, Andrea Bartolini, Luca Benini, Michela Milano
1598 - 1599
Scheduling and dispatching are critical enabling technologies in supercomputing and grid computing. In these contexts, scalability is an issue: we have to allocate and schedule up to tens of thousands of tasks on tens of thousands of resources. This problem scale is out of reach for complete and centralized scheduling approaches.
We propose a distributed allocation and scheduling paradigm called DARDIS that is lightweight, scalable and fully customizable in many domains. In DARDIS each task offloads to the available resources the computation of a probability index associated with each possible start time for the given task on the specific resource. The task then selects the proper resource and start time on the basis of the above probability.
Fabrizio Riguzzi, Elena Bellodi, Riccardo Zese, Giuseppe Cota, Evelina Lamma
1602 - 1603
Probabilistic Logic Programming is a promising formalism for dealing with uncertainty. Learning probabilistic logic programs has been receiving an increasing attention in Inductive Logic Programming: for instancethe system SLIPCOVER learns high quality theories in a variety of domains. HoweverSLIPCOVER is computationally expensivewith a running time of the order of hours. In order to apply SLIPCOVER to Big Data, we present SEMPRE, for “Structure lEarning by MaPREduce”, that scales SLIPCOVER by following a MapReduce strategy, directly implemented with the Message Passing Interface.
This paper proposes, for the first time in the literature, the use of hypergraphs for the efficient formation of effective coalitions. We put forward several formation methods that build on existing hypergraph algorithms, and exploit hypergraph structure to identify agents with desirable characteristics. Our approach allows the near-instantaneous formation of high quality coalitions, while adhering to multiple stated requirements regarding coalition quality. Moreover, our methods are shown to scale to dozens of thousands of agents within fractions of a second; with one of them scaling to even millions of agents within seconds. We apply our approach to the problem of forming coalitions to provide (electric) vehicle-to-grid (V2G) services. Ours is the first approach able to deal with large-scale, realtime coalition formation for the V2G problem, while taking multiple criteria into account for creating electric vehicle coalitions.
In coalition formation with self-interested agents both social welfare of the multi-agent system and stability of individual coalitions must be taken into account. However, in large-scale systems with thousands of agents, finding an optimal solution with respect to both metrics is infeasible.
In this paper we propose an approach for finding coalition structures with suboptimal social welfare and coalition stability in large-scale multi-agent systems. Our approach uses multi-agent simulation to model a dynamic coalition formation process. Agents are allowed to deviate from unstable coalitions, thus increasing the coalition stability. Furthermore we present an approach for estimating coalition stability, which alleviates exponential complexity of coalition stability computation. This approach is used for estimating stability of multiple coalition structures generated by the multi-agent simulation, which enables us to select a solution with high values of both social welfare and coalition stability. We experimentally show that our algorithms cause a major increase in coalition stability compared to a baseline social welfare-maximizing algorithm, while maintaining a very small decrease in social welfare.
Ecological Momentary Assessment (EMA) data is organized in multiple levels (per-subject, per-day, etc.) and this particular structure should be taken into account in machine learning algorithms used in EMA like decision trees and its variants. We propose a new algorithm called BBT (standing for Bagged Boosted Trees) that is enhanced by a over/under sampling method and can provide better estimates for the conditional class probability function. Experimental results on a real-world dataset show that BBT can benefit EMA data classification and performance.
This paper proposes a computational model based on peer reviews for assessing the reputation of researchers and research work. We argue that by relying on peer opinions, we address some of the pitfalls of current approaches for calculating the reputation of authors and papers. We also introduce a much needed feature for review management: calculating the reputation of reviews and reviewers.
Mohammed Hasanuzzaman, Wai Leung Sze, Mahammad Parvez Salim, Gaël Dias
1616 - 1617
Web search query logs can be used to track and, in some cases, anticipate the dynamics of individual behavior which is the smallest building block of the economy. We study AOL query logs and introduce a collective future intent index to measure the degree to which Internet users seek more information about the future than the past and the present. We have asked the question whether there is link between the collective future intent index and financial market fluctuations on a weekly time scale, and found a clear indication that the weekly transaction volume of S&P 500 index is correlated with the collective intent of the public to look forward.
We develop and test a new classification model learning algorithm that relies on the soft-label information and that is able to learn classification models more rapidly and with a smaller number of labeled instances than existing approaches.
Ashiqur R. KhudaBukhsh, Peter J. Jansen, Jaime G. Carbonell
1620 - 1621
Human experts or autonomous agents in a referral network must decide whether to accept a task or refer to a more appropriate expert, and if so to whom. In order for the referral network to improve over time, the experts must learn to estimate the topical expertise of other experts. This paper extends concepts from Reinforcement Learning and Active Learning to referral networks, to learn how to refer at the network level, based on the proposed distributed interval estimation learning (DIEL) algorithm. Diverse Monte Carlo simulations reveal that DIEL improves network performance significantly over both greedy and Q-learning baselines , approaching optimal given enough data.
Automatic short answer grading (ASAG) is the task of automatically grading students answers which are a few words to a few sentences long. While supervised machine learning techniques (classification, regression) have been successfully applied for ASAG, they suffer from the constant need of instructor graded answers as labelled data. In this paper, we propose a transfer learning based technique for ASAG built on an ensemble of text classifier of student answers and a classifier using numeric features derived from various similarity measures with respect to instructor provided model answers. We present preliminary empirical results to demonstrate efficacy of the proposed technique.
In this paper, we present a feature learning method for long-time sensor data. Although feature learning methods have been successfully used in many applications, they cannot extract features efficiently when the dimension of training data is quite large. To address this problem, we propose a method to search effective features from long-time sensor data. The important characteristic of our method is that it searches the features based on the gradient of input vectors to minimize the objective function of the learning algorithm. We apply our method to the estimation of physical capacity from wearable sensor data. The experimental results show that our method can estimate leg muscle strength more accurately than conventional methods using a feature learning method and current clinical index.
Fei Huang, Yong Cheng, Cheng Jin, Yuejie Zhang, Tao Zhang
1626 - 1627
In this paper we aim to employ deep learning to enhance SBIR via deep discriminative representation. Our main contributions focus on: 1) The deep discriminative representation is established to bridge both the visual appearance gap and the semantic gap between sketches and images; 2) The deep learning pattern is applied to our SBIR model through training on our transformed sketch-like images to overcome the rarity of training sketches. Our experiments on a large number of public sketch and image data have obtained very positive results.
Auke J. Wiggers, Frans A. Oliehoek, Diederik M. Roijers
1628 - 1629
In this paper, we introduce a new formulation for the value function of a zero-sum Partially Observable Stochastic Game (zs-POSG) in terms of a ‘plan-time sufficient statistic’, a distribution over joint sets of information. We prove that this value function exhibits concavity and convexity with respect to appropriately chosen subspaces of the statistic space. We anticipate that this result is a key pre-cursor for developing solution methods that exploit such structure. Finally, we show that the formulation allow us to reduce a finite zs-POSG to a ‘centralized’ model with shared observations, thereby transferring results for the latter (narrower) class of games to games with individual observations.
We propose an extension of the standard game description language for general game playing to include epistemic games, which are characterised by rules that depend on the knowledge of players. A single additional keyword suffices to define GDL-III, a general description language for games with imperfect information and introspection. We present an Answer Set Program for automatically reasoning about GDL-III games. Our extended language along with a suitable basic reasoning system can also be used to formalise and solve general epistemic puzzles.
For many applications it is important to be able to detect what a human is currently doing. This ability is useful for applications such as surveillance, human computer interfaces, games and health-care. In order to recognize a human action, the typical approach is to use manually labeled data to perform supervised training. This paper aims to compare the performance of several supervised classifiers trained with manually labeled data versus the same classifiers trained with data automatically labeled. In this paper we propose a framework capable of recognizing human actions using supervised classifiers trained with automatically labeled data in RGB-D videos.
Mohammed Hasanuzzaman, Gaël Dias, Stéphane Ferrari
1634 - 1635
The ability to capture the time information conveyed in natural language is essential to many natural language processing applications such as information retrieval, question answering, automatic summarization, targeted marketing, loan repayment forecasting, and understanding economic patterns. In this paper, we propose a graph-based semi-supervised classification strategy that makes use of WordNet definitions or ‘glosses’, its conceptual-semantic and lexical relations to supplement WordNet entries with information on the temporality of its word senses. Intrinsic evaluation results show that the proposed approach outperforms prior semi-supervised, non-graph classification approaches to the temporality recognition of word senses, and confirm the soundness of the proposed approach.
Although recent studies have shown that a Bayesian network classifier (BNC) that maximizes the classification accuracy (i.e., minimizes the 0/1 loss function) is a powerful tool in knowledge representation and classification, this classifier focuses on the majority class, is usually uninformative about the distribution of misclassifications, and is insensitive to error severity (making no distinction between misclassification types). We propose to learn a BNC using an information measure (IM) that jointly maximizes classification and information, and evaluate this measure using various databases. We show that an IM-based BNC is superior to BNCs learned using other measures, especially for ordinal classification and imbalanced problems, and does not fall behind state-of-the-art algorithms with respect to accuracy and amount of information provided.