196,616 research outputs found
Analysis and Optimal Management of Mobile Networks Towards Pervasive Intelligence
Full text linkQuesta tesi indaga le tecniche di machine learning come potenti strumenti per l'analisi e la gestione ottima delle reti radiomobili per ottenere l’Intelligenza Pervasiva.
In questo contesto, descrive soluzioni recenti progettate per affrontare tre problemi diversi, ma correlati.
In primo luogo, dopo aver adattato una metodologia di unsupervised learning per caratterizzare i pattern di utilizzo delle risorse radio, un modello di multi-task learning, eseguito direttamente ai bordi della rete, è ideato per eseguire il data mining dal canale di controllo di una rete radiomobile operativa.
Due configurazioni di reti neurali, basate su autoencoder Undercomplete o Sequence to Sequence, vengono sfruttate per ottenere rappresentazioni di feature comuni dei profili di traffico.
In seguito, i softmax e fully-connected layer vengono utilizzati rispettivamente per anticipare le informazioni sul tipo di traffico da servire e il pattern di utilizzo delle risorse radio richiesto da ciascun servizio durante la sua esecuzione.
Inoltre, viene sfruttato un approccio Software-Defined Networking per monitorare la mobilità degli utenti.
Di conseguenza, la predizione sia della distribuzione degli utenti tra le celle che delle risorse di comunicazione e di computazione richieste su diversi orizzonti temporali futuri viene eseguita attraverso un'architettura Convolutional Long Short-Term Memory.
Queste informazioni vengono utilizzate per eseguire in anticipo l'allocazione e la distribuzione delle richieste degli utenti tramite Dynamic Programming.
Infine, viene proposto uno schema Tenant-driven di slicing enforcement nella rete di accesso radio basato sull'Intelligenza Pervasiva per evitare l'over-provisioning delle risorse radio risparmiando la banda (ovvero, il paradigma Pay for What You Get).
L’Infrastructure Provider utilizza un autoencoder convoluzionale per comprimere le informazioni sulle risorse di rete e sulla connettività e le condivide con i Tenant.
A sua volta, ogni Tenant implementa un algoritmo Deep Deterministic Policy Gradient per adattare dinamicamente le richieste di banda in base alle esigenze dei propri utenti.
I risultati di questo algoritmo vengono quindi utilizzati dall’Infrastructure Provider per applicare efficacemente lo slicing della rete.
L'indagine in scenari reali e il confronto con approcci convenzionali adottati per l'analisi e la gestione ottima delle reti radiomobili dimostrano l'efficacia delle soluzioni proposte basate sul machine learning. In termini di applicabilità, le metodologie progettate sono anche in linea con l'evoluzione delle reti radiomobili, in cui l'Intelligenza Artificiale sarà integrata in modo nativo e pervasivo per consentire la piena automazione della rete.This thesis investigates machine learning techniques as powerful tools for the analysis and the optimal management of mobile networks towards Pervasive Intelligence.
In this context, it describes recent solutions conceived for addressing three different, but related, issues.
Firstly, after tailoring an unsupervised learning methodology to characterize radio resource utilization patterns, a Multi-Task Learning model, running directly at the edge of the network, is conceived to perform data mining from the control channel of an operative mobile network.
Two configurations of neural networks, based on Undercomplete or Sequence to Sequence autoencoders, are exploited to obtain common feature representations of traffic profiles.
Then, softmax and fully-connected layers are used to anticipate information on the type of traffic to be served and the radio resource utilization pattern requested by each service during its execution, respectively.
Moreover, a Software-Defined Networking approach is exploited to monitor users’ mobility.
Consequently, the prediction of both the distribution of users among cells and the communication and computational resources they request over different look-ahead horizons is performed through a Convolutional Long Short-Term Memory architecture.
This information is used to perform anticipatory allocation and distribution of users' requests via Dynamic Programming.
Hence, a Tenant-driven Radio Access Network slicing enforcement scheme based on Pervasive Intelligence is proposed to avoid the radio resources over-provisioning while saving bandwidth (i.e., the Pay for What You Get paradigm).
The Infrastructure Provider exploits a convolutional autoencoder to compress the information on network resources and connectivity and share it with the Tenants.
In turn, each Tenant implements a Deep Deterministic Policy Gradient algorithm to dynamically adapt bandwidth requests according to its own users' requirements.
The outcomes of this algorithm are then used by the Infrastructure Provider to effectively enforce network slicing.
The investigation in real scenarios and the comparison against conventional approaches adopted for the analysis and the optimal management of mobile networks demonstrate the effectiveness of the proposed machine learning-based solutions. In terms of applicability, the conceived methodologies are also in line with the evolution of mobile networks, where Artificial Intelligence will be natively and pervasively integrated for enabling full network automation
The Relation Between the Internationalisation of Services and the Process of Innovation: a research agenda
No full textWhen do combinatorial mechanisms apply in the production of inflected words?
No full textCholin J, Rapp B, Miozzo M. When do combinatorial mechanisms apply in the production of inflected words? Cognitive Neuropsychology. 2010;27(4):334-359
Significance of clustered tumor suppressor genes in cancer
No full textEvaluation of: Xue W, Kitzing T, Roessler S et al. A cluster of cooperating tumor-suppressor gene candidates in chromosomal deletions. Proc. Natl Acad. Sci. USA 109, 8212-8217 (2012). The two-hit model is a well-known mechanism for the inactivation of tumor suppressor genes in cancer and it has been assumed that chromosomal deletions are the second inactivating event. Large deletions are frequently found in cancer and can lead to the haploinsufficiency of the loci mapped to the deleted region. The study by Xue et al. demonstrated that hemizygous 8p deletions can attenuate the activity of multiple genes that control growth and promote tumorigenesis, and showed that the effect of large 8p deletions on tumor phenotype goes beyond the effects of the individual genes as the characteristics of a tumor are also influenced by the additive and/or combined effect of the haploinsufficiency of multiple genes. These convincing findings, demonstrating that the hemizygosity of a cluster of genes negatively regulates proliferation and promotes tumor growth, have opened up new study perspectives aimed at characterizing the genomic organization of this new class of tumor suppressor genes and their role in tumorigenesis
The Relation Between the Internationalisation of Services and the Process of Innovation: a research agenda
No full textThe role of imprinted genes in fetal growth
No full textGenomic imprinting is the phenomenon by which one of the two alleles of a subset of genes is preferentially expressed according to its parental origin. This pattern of inheritance is different from the more frequent mode of Mendelian inheritance, which is not influenced by the parental origin of the allele. The idea that imprinted genes can affect fetal growth is becoming increasingly intriguing as it has been shown that most imprinted genes are expressed in the placenta and some play a role in regulating the interactions between its fetal and maternal interfaces. This article considers genomic imprinting by reviewing recent findings of alterations in fetal growth related to different types of genetic changes affecting the expression of imprinted genes. Among the genetic anomalies, the uniparental disomy (UPD) defines the inheritance of both homologous chromosomes from only one parent. UPDs of a number of chromosomes have been described in association with effects on the phenotype. We reviewed cases of UPD reported till now with particular reference to those associated to growth alterations
Evidence for morpho-phonological processes in spoken production
No full textGoldberg AM, Cholin J, Bertz JW, Rapp B, Miozzo M. Evidence for morpho-phonological processes in spoken production. Brain and Language. 2007;103:162-163
Modularity and innovation in knowledge-intensive business services: IT outsourcing in Germany and the UK
No full textDrawing on an empirical study of IT outsourcing in the UK and Germany, this paper explores the lessons for modularity that can be drawn from the outsourcing of knowledge-intensive business services (KIBS). Because of the inseparability of information and production technologies, IT outsourcing is frequently accompanied by wider transformations in clients' production technologies. This results in the need for knowledge and organisational coordination in the form of the transfer of staff from the client and the retained IT organisation. Modularity is often presented as a design strategy that stimulates innovation. Our research findings challenge the generalisability of this claim when examining KIBS outsourcing. We show that intangibility of services exacerbates the conflicts between clients and suppliers, which may present obstacles to innovation. © 2005 Elsevier B.V. All rights reserved
Institutional effects on the IT outsourcing market: Analysing clients, suppliers and staff transfer in Germany and the UK
No full textDrawing on empirical evidence in Germany and the UK, this article examines the institutional effects on a fast-growing area of knowledge-intensive business services - IT outsourcing. This is an important area for research since the IT outsourcing market provides many organizations with an important specialized production input and is characterized by complex inter-organizational relations. By exploring institutional influences in the context of IT outsourcing, the research extends earlier studies on how client-supplier relations shape markets for business services. It also contributes to varieties of capitalism debates by highlighting heterogeneous institutional effects within countries and common systemic trends (involving powerful multinational IT firms) in the development of the market for IT outsourcing. Comparative analysis of 13 IT outsourcing contracts in Germany and the UK, focusing on the organizational practices of client organizations and IT firms, illuminated institutional effects within the organizational setting. Analysis of industrylevel data shows that the diverse institutional contexts of Germany and the UK provided an equally favourable basis for growth in the IT outsourcing market, despite its apparent deregulatory bias. But significant institutional effects were observed, specifically related to: the role of deliberative institutions (especially works councils); and institutions governing technical standards and contracting rules. Strong deliberative institutions in Germany facilitated market growth since transactions involved distributive dilemmas, particularly related to staff transfer. Also, while institutions shaped technical and contractual expertise of client managers, they were not deterministic. Instead, they interacted with characteristics of the IT outsourcing market, namely: heterogeneous client practices to improve absorptive capacity; public vs. private contracting experience; and power relations between client and IT firm in their use of market discipline. Copyright © 2006 SAGE Publications
- …
