AITIA International Inc.

Budapest, Hungary

AITIA International Inc.

Budapest, Hungary
Time filter
Source Type

Gulyas L.,PetaByte Ltd | Gulyas L.,AITIA International Inc | Gulyas L.,Eötvös Loránd University | Kampis G.,PetaByte Ltd | And 2 more authors.
European Physical Journal: Special Topics | Year: 2013

We develop and analyze an agent-based model for the study of information propagation in dynamic contact networks. We represent information as a state of a node in a network that can be probabilistically transferred to an adjacent node within a single time step. The model is based on a closed (yet sufficiently large) population that can support processes of link generation and annihilation using different contact regimes. Our study is confined to the case of homogeneous contacts, where each agent establishes and breaks contacts in the same way. We consider information to be available for spreading in a fixed time window (i.e. finite memory). We find, surprisingly, that information transmission (measured as the proportion of informed nodes after a fixed number of time steps) is identical for dynamic preferential and random networks, but radically different for the associate mixing contact regime. We also find that the probability of transmission is, similarly counterintuitively, not a main driver of the process as opposed the the main network par maters determining contact lifetime and the turnover rate on connections. We discuss the explanation and the significance of these results in the light of the fundamental difference between dynamic and static (cumulative) networks. © 2013 EDP Sciences and Springer.

Plosz S.,AIT Austrian Institute of Technology | Hegedus C.,AITIA International Inc. | Varga P.,Budapest University of Technology and Economics
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2016

The Arrowhead Framework aims to create collaborative automation using networked embedded devices by establishing a service oriented approach to govern them. Various cyber-physical Systems can provide and consume Services from one another in closed automation clouds. These System-of-Systems has been introduced by the Arrowhead framework as Local Clouds. These clouds – being high value targets – can then be subject to an extensive amount of threats. This paper is dedicated towards revising the Arrowhead framework to further enhance its security solutions. A certificate-based architecture is presented to solve authentication and authorization tasks not just within, but in-between Local Clouds by using a token concept applied for services. This schema also allows the integration of resource constrained devices in coexistence with different levels of security. © Springer International Publishing Switzerland 2016.

Mihajlik P.,Budapest University of Technology and Economics | Mihajlik P.,ech Research Center Nonprofit | Tuske Z.,Budapest University of Technology and Economics | Tarjan B.,Budapest University of Technology and Economics | And 3 more authors.
IEEE Transactions on Audio, Speech and Language Processing | Year: 2010

Various morphological and acoustic modeling techniques are evaluated on a less resourced, spontaneous Hungarian large-vocabulary continuous speech recognition (LVCSR) task. Among morphologically rich languages, Hungarian is known for its agglutinative, inflective nature that increases the data sparseness caused by a relatively small training database. Although Hungarian spelling is considered as simple phonological, a large part of the corpus is covered by words pronounced in multiple, phonemically different ways. Data-driven and language specific knowledge supported vocabulary decomposition methods are investigated in combination with phoneme- and grapheme-based acoustic modeling techniques on the given task. Word baseline and morph-based advanced baseline results are significantly outperformed by using both statistical and grammatical vocabulary decomposition methods. Although the discussed morph-based techniques recognize a significant amount of out of vocabulary words, the improvements are due not to this fact but to the reduction of insertion errors. Applying grapheme-based acoustic models instead of phoneme-based models causes no severe recognition performance deteriorations. Moreover, a fully data-driven acoustic modeling technique along with a statistical morphological modeling approach provides the best performance on the most difficult test set. The overall best speech recognition performance is obtained by using a novel word to morph decomposition technique that combines grammatical and unsupervised statistical segmentation algorithms. The improvement achieved by the proposed technique is stable across acoustic modeling approaches and larger with speaker adaptation. © 2010 IEEE.

Panica S.,West University of Timișoara | Petcu D.,West University of Timișoara | Larrate I.L.,Tecnalia | Mahr T.,AITIA International Inc.
Proceedings - 2012 11th International Symposium on Parallel and Distributed Computing, ISPDC 2012 | Year: 2012

The elasticity and scalability promised by the Cloud computing paradigm, intended mainly for the new Web 2.0 applications, attracts also legacy applications owners. The migration of legacy applications towards Cloud is a complex process and only few tools and platforms are available to assist the applications' owners in this process. This paper explains the challenges that arise in the migration process of legacy applications and also exposes the solution offered by an open-source platform as a service, named mOSAIC, designed to deal with applications portability between Clouds. © 2012 IEEE.

Di Martino B.,The Second University of Naples | Petcu D.,Institute E Austria Timisoara | Cossu R.,European Space Agency | Goncalves P.,Terradue SRL | And 2 more authors.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2011

The current diversity of Cloud computing services, benefic for the fast development of a new IT market, hinders the easy development, portability and inter-operability of Cloud oriented applications. Developing an application oriented view of Cloud services instead the current provider ones can lead to a step forward in the adoption of Cloud computing on a larger scale than the actual one. In this context, we present a position paper exposing the concepts behind a recent proposal for an open-source application programming interface and platform for dealing with multiple Cloud computing offers. © 2011 Springer-Verlag Berlin Heidelberg.

Dugundji E.R.,University of Amsterdam | Gulyas L.,AITIA International Inc.
Studies in Computational Intelligence | Year: 2013

The reported research treats interactions between agents and generated feedback dynamics in the adoption of various transportation mode alternatives. We consider a simple nested logit model where an agent's choice is directly influenced by the percentages of the agent's neighbors and socio-economic peers making each choice, and which accounts for common unobserved attributes of the choice alternatives in the error structure. We explicitly address non-global interactions within several hypothesized social and spatial network structures. Discrete choice estimation results controlling heterogeneous individual preferences are embedded in a multi-agent based simulation model in order to observe the evolution of choice behavior over time with socio-dynamic feedback due to the network effects. For the particular simple model under study, we find the impact of initial conditions on the emergent long-run behavioral outcomes is dependent on network size and network connectivity. We conclude highlighting limitations of our present study and recommendations for future work. © 2013 Springer-Verlag Berlin Heidelberg.

Dugundji E.R.,University of Amsterdam | Gulyas L.,AITIA International Inc.
Procedia Computer Science | Year: 2012

We consider a model where a commuter's transportation mode choice is directly influenced by percentages of neighbors and socioeconomic peers making each choice. Discrete-choice estimation results controlling overall mechanisms related to individual heterogeneous preferences are embedded in a multi-agent based simulation in order to observe the evolution of choice behavior over time with sociodynamic feedback. We find that the estimated utility parameters for different plausible sociogeographic network scenarios can generate dramatically different dynamics. However, in a hypothetical experiment we find that swapping sociogeographic networks does not significantly change the long-run outcome of the simulation, when utility parameters are held fixed. © 2012 Published by Elsevier Ltd.

Legendi R.O.,AITIA International Inc. | Legendi R.O.,Eötvös Loránd University | Gulyas L.,AITIA International Inc. | Gulyas L.,Eötvös Loránd University
Advances in Intelligent Systems and Computing | Year: 2014

Inspecting the dynamics of networks opens a new dimension in understanding the mechanisms behind complex social phenomena. In our previous work, we defined a set of elementary dynamic models based on classic random and preferential networks. Focusing on edge dynamics, we defined several processes for changing networks with a fixed set of vertices. We applied simple rules, including the combination of random, preferential and assortative mixing of existing edges. Starting from an empty initial network, we examined network properties (like density, clustering, average path length, number of components and degree distribution) of both snapshot and cumulative networks for various lengths of aggregation time windows. In this paper, we extend our analysis with a comparison to results obtained from empirical data from two selected data sets. The knowledge of the baseline behavior of the abstract elementary dynamic network models helps us to identify the important, idiosyncratic properties of the empirical networks. © Springer-Verlag Berlin Heidelberg 2014.

Rak M.,The Second University of Naples | Venticinque S.,The Second University of Naples | Mahr T.,AITIA International Inc. | Echevarria G.,Tecnalia | Esnal G.,Tecnalia
Proceedings - 2011 3rd IEEE International Conference on Cloud Computing Technology and Science, CloudCom 2011 | Year: 2011

Cloud computing delegates the management of any kind of resources, such as the computing environment or storage systems for example, to the network. The wide-spread permeation of the Cloud paradigm implies the need of new programming models that are able to utilize such new features. Once the problem of enabling developers to manage Cloud resources in a clear and flexible way is solved, a new problem emerges: the monitoring of the quality of the acquired resources and of the services offered to final users. As the first step, the mOSAIC API and framework aim at offering a solution for the development of interoperable, portable and Cloud-provider independent Cloud applications. As the second step, this paper introduces the mOSAIC monitoring components that facilitate the building of custom monitoring systems for Cloud applications using the mOSAIC API. © 2011 IEEE.

Dugundji E.R.,University of Amsterdam | Gulyas L.,AITIA International Inc.
Computational and Mathematical Organization Theory | Year: 2013

Suppose you have the possibility to choose to adopt one of a number of different behaviors or to choose to buy one of a number of different products, and suppose your choice is influenced by your individual perception of the average choices made by others. Economists Brock and Durlauf (in Am. Econ. Rev. 92(2):298, 2002; The Economy as an Evolving Complex System III. Oxford University Press, New York, 2006) have derived seminal theoretical results for the equilibrium behavior of the multinomial discrete choice model with social interactions, assuming homogeneous decision-makers, global interactions and laws of large of numbers. The research presented in this paper extends Brock and Durlauf's model to allow for unobserved preference heterogeneity between choice alternatives by studying the nested logit model. Next, by drawing on the computational possibilities permitted through social simulation of multi-agent systems (MAS), this paper relaxes the assumption of global interactions and considers instead local interactions within several hypothesized social and spatial network structures. Additional heterogeneity is thus hereby induced by the influence on a given decision-maker's choice by the particular network connections he or she has and the particular perceived percentages, for example, of the agent's neighbors or socio-economic peers making each choice. Discrete choice estimation results controlling these heterogeneous individual preferences are embedded in a multi-agent based simulation model in order to observe the evolution of choice behavior over time with socio-dynamic feedback due to the network effects. The MAS approach also gives us an additional advantage in the possibility to test size effects, and thus relax the assumption of large numbers, as well as test the effect of different initial conditions. Finally an extra benefit is gained via the MAS approach in that we are not confined to study only the equilibrium behavior, and have the possibility here to observe the time-varying trajectories of the choice behavior. This is important since smaller network sizes are revealed to be associated with higher volatility of the choice behavior in this model, and consequently stochastic cycling between equilibria. Averaged over time, the emergent behavior in such case yields a quite different picture than the theoretical results predicted by Brock and Durlauf. Furthermore being able to observe the emergent behavior allows us to see the subtle role of the unobserved heterogeneity in the nested logit model in breaking the symmetry of the multinomial logit model. We can see the temporal patterns by which theoretically predicted dominant equilibria emerge or not according to different social and spatial network scenarios. With an eye towards application in the context of transportation mode choice, we conclude highlighting limitations of our present study and recommendations for future work. © 2013 Springer Science+Business Media New York.

Loading AITIA International Inc. collaborators
Loading AITIA International Inc. collaborators