http://old.hiit.fi/events/2007 en http://old.hiit.fi/node/434 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even">HIIT researcher Merja Oja presents her doctoral thesis <em>Methods for Exploring Genomic Data Sets: Application to Human Endogenous Retroviruses</em> in a public dissertation at the Helsinki University of Technology. The custos of the event is professor Samuel Kaski. The opponent is Professor Hiroshi Mamitsuka, Kyoto University, Japan. [<a href="http://www.tkk.fi/nyt/vaitoksia/vaitostiedotteet_2007/Oja_Merja.pdf">v&auml;it&ouml;stiedote</a>]</div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Fri, 14 Dec 2007 10:00:00 +0000 mam 434 at http://old.hiit.fi http://old.hiit.fi/node/408 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even">HIIT seminars in fall 2007 will be held in hall **B222** of Exactum,<br /> on Fridays starting at 10:15 a.m. Coffee available from 10.<br /> <br /> Fri Dec 14<br /> Patrik Hoyer<br /> <p>Exploiting both non-gaussianity and conditional independencies for causal discovery<br /> <br /> Abstract:</p> <p>I will describe my recent visit to Carnegie Mellon University in Pittsburgh, both in terms of the general experience and in terms of the scientific results. In particular, I will describe techniques for learning linear causal models that combine the strengths of their methods based on conditional independencies and our algorithms based on non-gaussianity.</p> <p>This is joint work with Peter Spirtes, Richard Scheines, Clark Glymour, Joseph Ramsey and Gustavo Lacerda at Carnegie Mellon University, Shohei Shimizu at the Institute of Statistical Mathematics in Tokyo, and Aapo Hyv&auml;rinen at HIIT.</p></div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Fri, 14 Dec 2007 08:15:00 +0000 tkujala 408 at http://old.hiit.fi http://old.hiit.fi/node/415 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even">HIIT's Christmas party will be organized on Wednesday 12 December at 5 p.m. The place is Hotel Katajanokka, Vy&ouml;katu 1.</div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Wed, 12 Dec 2007 15:00:00 +0000 tkujala 415 at http://old.hiit.fi http://old.hiit.fi/node/407 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even">HIIT seminars in fall 2007 will be held in hall **B222** of Exactum,<br /> on Fridays starting at 10:15 a.m. Coffee available from 10.<br /> <br /> Fri Dec 7<br /> <p>Petteri Kaski<br /> Enumeration of the 1-factorizations of the complete graph K_{14}<br /> <br /> Abstract:<br /> In how many different ways can one organize a round-robin football tournament for 14 teams? Two answers conditioned on the precise definition of ``different'' are 98758655816833727741338583040 and 1132835421602062347. This talk provides some background on the problem and discusses the combinatorial algorithms used to compute these numbers.<br /> <br /> This is joint work with Patric &Ouml;sterg&aring;rd.</p></div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Fri, 07 Dec 2007 08:15:00 +0000 tkujala 407 at http://old.hiit.fi http://old.hiit.fi/node/406 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even">HIIT seminars in fall 2007 will be held in hall **B222** of Exactum,<br /> on Fridays starting at 10:15 a.m. Coffee available from 10.<br /> <br /> Fri Nov 30<br /> Jarkko Tikka<br /> Two approaches to select relevant inputs for RBF network<br /> <p>Abstract:</p> <p>In regression problems, making accurate predictions is often the primary goal. Neural networks provide good generalization in many cases, but their interpretability is usually limited. The dependencies between input<br /> and output variables is not clear at all. However, the contributions of input variables in the prediction of output would be valuable information in many real world applications. Two input selection algorithms for Radial Basis Function (RBF) networks are presented to increase the interpretability of the models. Firstly, a sequential input selection algorithm is proposed. It is a backward selection type algorithm based on the partial derivatives of the RBF network. Secondly, a input selection method, which is based on a constrained cost function, is shown. The input selection algorithms are applied to both simulated and benchmark data sets and the results are convincing. That is, both algorithms yield accurate prediction models, which are also parsimonious in terms of input variables.</p></div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Fri, 30 Nov 2007 08:15:00 +0000 tkujala 406 at http://old.hiit.fi http://old.hiit.fi/node/359 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even">HIIT Board meeting 5/2007 has been cancelled.</div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Tue, 27 Nov 2007 22:00:00 +0000 mam 359 at http://old.hiit.fi http://old.hiit.fi/node/277 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even">HIIT seminars in fall 2007 will be held in hall **B222** of Exactum,<br /> on Fridays starting at 10:15 a.m. Coffee available from 10.<br /> <br /> Fri Nov 23<br /> Kai Nordlund<br /> Accelerator Laboratory, Department of Physical Sciences <br /> <br /> Computational physics and chemistry at the Kumpula Campus<br /> <br /> Abstract:<br /> I will first present what the role of computer simulations are in physics, namely how they relate to theoretical and experimental physics. I will then give an overview of computational physics and chemistry activities on our campus, showing that simulations have an important role in almost all active research areas. To give an idea of what computer simulations of processes occuring in nature may mean in practise, I present three central algorithms used widely in different branches of physics and chemistry: Metropolis Monte Carlo, Kinetic Monte Carlo and Molecular dynamics. I also point out that especially the first of these is also widely used in mathematics and computer science, even though it was originally devised to simulate a problem very much belonging to physics only, the melting of a hard sphere system. Finally I will present some results of recent research in our group on how molecular dynamics can be used to understand how the properties of a new nanomaterial, carbon nanotubes, can be improved on by ion irradiation.</div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Fri, 23 Nov 2007 08:15:00 +0000 tkujala 277 at http://old.hiit.fi http://old.hiit.fi/node/278 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even">HIIT seminars in fall 2007 will be held in hall **B222** of Exactum,<br /> on Fridays starting at 10:15 a.m. Coffee available from 10.<br /> <br /> Fri Nov 16<br /> Simo Vanni<br /> Brain Research Unit, Helsinki University of Technology<br /> <br /> Systems neuroscience approach to visual information processing<br /> <br /> Abstract:<br /> Biological vision can recognise objects and comprehend space in a fraction of a second, but the mechanisms behind this efficiency are still unclear. Strong research effort for decades is now converging towards models, which may be able to show where to look for such efficiency. Because the mechanisms are unknown, we need to explore. Systems neuroscience is a multidisciplinary field, which combines various methods to study population behaviour of neurons. From the non-invasive methods, functional magnetic resonance imaging has the best spatial resolution, and may be used for mapping of active and inactive areas in the brain, quantification of local signals, and analysis of spatial or temporal correlations in the different neural populations. Modifications of visual stimuli or analysis of behavioural measures can be combined to statistical analysis of fMRI results. This talk introduces our recent fMRI results suggesting new views to organizational principles of human visual cortex.</div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Fri, 16 Nov 2007 08:15:00 +0000 tkujala 278 at http://old.hiit.fi http://old.hiit.fi/node/412 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even">Dimitris Mavroeidis from Athens University of Economics and Business, <br /> Greece, will give second talk<br /> <br /> Wednesday Nov 14, 10:15 o'clock, in C222:<br /> <br /> Stability based Sparse LSI/PCA, with extensions to K-Means and Spectral Clustering. <br /> <br /> Abstract:<br /> <br /> The stability of sample based algorithms is a concept commonly used for parameter tuning and validity assessment. In this presentation we focus on two well studied algorithms, LSI and PCA, and propose a feature selection process that provably guarantees the stability of their outputs. The feature selection process is performed such that the level of (statistical) accuracy of the LSI/PCA input matrices is adequate for computing meaningful (stable) eigenvectors. The feature selection process ``sparsifies&quot; LSI/PCA, resulting in the projection of the instances on the eigenvectors of a principal submatrix of the original input matrix, thus producing sparse factor loadings that are linear combinations solely of the selected features. We utilize bootstrapping confidence intervals for assessing the statistical accuracy of the input sample matrices, and matrix perturbation theory in order to relate the statistical accuracy to the stability of eigenvectors. Finally, we demonstrate the proposed methodological approach can be extended to handle K-Means and Spectral Clustering.<br /> <br /> http://www.db-net.aueb.gr/index.php/corporate/people/phd_students/dimitrios_mavroeidis</div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Wed, 14 Nov 2007 08:15:00 +0000 tkujala 412 at http://old.hiit.fi http://old.hiit.fi/node/411 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even">Dimitris Mavroeidis from Athens University of Economics and Business, <br /> Greece, will give first talk<br /> <br /> Monday 12 Nov, 13:00 o'clock, in C221:<br /> <br /> Exploiting Semantic Thesauri in Text Classification and Information Retrieval using SVMs. <br /> <br /> Abstract:<br /> <br /> The introduction of semantic thesauri (such as WordNet) that contain a significant amount of semantic information, has led researchers to investigate their potential use for improving performance of the text <br /> classification and Information Retrieval (IR) tasks, extending the traditional bag of words representation, incorporating syntactic and semantic relationships among words. In this talk, we will focus on the SVM <br /> paradigm and present recent results that successfully incorporate semantic thesauri in the Text Classification and (IR) domain. Motivated by the experimental results, we will also present possible extensions using <br /> Sparse SVMs.</div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Mon, 12 Nov 2007 11:00:00 +0000 tkujala 411 at http://old.hiit.fi http://old.hiit.fi/node/405 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even">HIIT seminars in fall 2007 will be held in hall **B222** of Exactum, on Fridays starting at 10:15 a.m. Coffee available from 10.<br /> <br /> Fri Nov 9<br /> Anna Pernest&aring;l<br /> KTH School of Electrical Engineering, Automatic Control<br /> <br /> Bayesian fault diagnosis -- utilizing data and prior information<br /> <br /> Abstract: <br /> Fault diagnosis is the task of detecting if there are faults present in a system, and if so determine which. Due to noise, limited information, and model errors it is often difficult, or even impossible, to exactly point out the faults that are present. Hence, we are forced to reason under uncertainty.<br /> <br /> In this presentation we show how Bayesian methods can be applied to perform fault diagnosis. In particular we discuss how training data can be used for learning. One challenge is that the amount of data available is often limited. Therefore we also discuss how to use prior information: which information that may be available, and how it could be integrated in the Bayesian framework for learning.<br /> <br /> The methods are applied to the diagnosis of an automotive diesel engine, and we will present some preliminary results, based on authentic data from real driving situations.</div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Fri, 09 Nov 2007 08:15:00 +0000 tkujala 405 at http://old.hiit.fi http://old.hiit.fi/node/401 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even">HIIT researcher, M.Sc. Oriana Riva will defend her doctoral dissertation <em>Middleware for Mobile Sensing Applications in Urban Environments</em> on Friday, November 2nd at 12 o'clock at the University of Helsinki Main Building, Auditorium XIV, Unioninkatu 34. The opponent is Prof. Olaf Dr&ouml;gehorn (University of Kassel) and the custos Prof. Kimmo Raatikainen (University of Helsinki).</div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Fri, 02 Nov 2007 10:00:00 +0000 mam 401 at http://old.hiit.fi http://old.hiit.fi/node/402 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even">This seminar has been cancelled.<br /> <p>&nbsp;</p></div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Fri, 02 Nov 2007 08:15:00 +0000 tkujala 402 at http://old.hiit.fi http://old.hiit.fi/node/400 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even"><p> HIIT seminars in fall 2007 will be held in hall **B222** of Exactum, on Fridays starting at 10:15 a.m. Coffee available from 10. </p> <p> Fri Oct 12<br /> Dr. Thore Husfeldt<br /> Computer Science Department, Lund University </p> <p> Inclusion-exclusion in combinatorial optimisation </p> <p> Abstract: <br /> I present the well-known principle of inclusion-exclusion and survey its applications to algorithms, namely to combinatorial optimisation. This includes classical results like Ryser's formula for the permanent, forgotten results for the Hamiltonian cycles, and recent results on partitioning and subset convolution, presented within the framework of partial order theory. The amibition is to make the presentation self-contained and accessible to an audience encumbered by not much more than an introductory algorithms class and some discrete maths. </p></div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Thu, 11 Oct 2007 21:00:00 +0000 tkujala 400 at http://old.hiit.fi http://old.hiit.fi/node/369 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even">&nbsp;Guest Lecture by professor Bijendra Jain<br /> <br /> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; (Indian Institute of Technology Delhi, India)<br /> <br /> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; on Friday 12th October at 14:15<br /> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Exactum, room C220<br /> <br /> <br /> Title:<br /> &quot;Tracking targets using sensor networks under energy constraints&quot;<br /> <br /> Abstract:<br /> In this talk we consider tracking a target moving in a 2D plane using a network of sensors, each of which is capable of detecting presence of the target in its vicinity and communicating data to a central tracker. Information collated by the central tracker is presented in the form of&nbsp; a sub-set $\Sigma^{t^k}$ of sensors in whose vicinity the target is estimated to be present at time $t^k$. Given that available energy in sensors is at a premium, we have proposed protocols for target&nbsp; detection and route activation that require sensors to conserve energy by switching between inactive and active modes of operations. These protocols have been evaluated from the viewpoint of delay in target detection (or in route activation) and duty cycle, the fraction of the time for which a sensor is awake in inactive mode. Using simulation results we argue that this approach allows one to track a moving object with reasonable accuracy, provided the density of sensors is adequately large.<br /> <br /> Speaker Profile:<br /> Professor Bijendra Jain obtained B. Tech. from IIT Kanpur in 1970, and Ph.D. from SUNY, Stony Brook in 1975, both in Electrical Engg. Since 1975 he has been with IIT Delhi, where he is currently Deputy Director (Faculty) and Professor of Computer Science. In the past he has held visiting assignments with Universities of Texas and Maryland, Bell Labs, and Cisco Systems.<br /> His interest is in Computer Networks and Systems, including network models and analysis, algorithms for large sparse matrix operations, scheduling algorithms for hard real-time systems, fault-tolerant routing. His recent interest is in ad hoc and sensor networks. His research is funded in part by Government of India, UNDP, US Army, Sun Microsystems, Microsoft and Media Lab Asia.<br /> As early as 1989, Prof. Jain, together with developers from other institutions in India, built and launched India's first data network, ERNet. Today, ERNet is also a thriving not-for-profit company connecting over a million users spread across over 2000 institutions in India. He is co-inventor in five US patents, covering speed-up in access to Web pages and efficient monitoring of IP network performance. <br /> He has co-authored "OSI: Its Architecture and Protocols", a book published by McGraw Hill, New York. He is an active industry consultant and a member of several Government committees, including Naval Research Board and Government's Information Security Panel.<br /> Finally, Professor Jain is a co-founder and past-Chairman of Kritikal Solutions, a technology start-up focused on computer vision, embedded systems and networks.<br /> <br /> WELCOME!<br /> <br /> Patrik Flor&eacute;en and Kimmo Raatikainen</div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Thu, 11 Oct 2007 21:00:00 +0000 tkujala 369 at http://old.hiit.fi http://old.hiit.fi/node/370 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even"><p> HIIT järjestää yhdessä CICERO Learning -verkoston tutkijapäivän 9.10.2007 klo 13.00-16.00 Exactumissa C222 tutustumis- ja verkottumistarkoituksessa. </p> <p> Alustava ohjelma </p> <p> Oppimisen tilat ja teknologiat </p> <p> 13.00 Tutkijapäivän avaus ja kahvia, varareht. Hannele Niemi<br> 13.05 CICERON esittely ja tutkimus<br> 13.45 HIITin esittely ja tutkimus, akat. prof Heikki Mannila ja tutkijat<br> 14.15 Tekesin Tila-ohjelman esittely, Satu Haaparanta, Tekes<br> 14.45 Tauko ja kahvia<br> 15.00 Mahdollinen tutkimusyhteistyö ja yhteistyömuodot, pj prof. Kristiina Kumpulainen<br> 16.00 Tutkijapäivä loppuu </p> <p> CICERO Learning on Helsingin yliopiston koordinoima suomalaisten yliopistojen oppimista ja osaamista tutkivien monitieteisten tutkimusryhmien vuonna 2005 perustettu verkosto. (Ks. <a href="http://www.cicero.fi">http://www.cicero.fi</a>) </p> <p> Tarkempi ohjelma ja ilmoittautuminen lähempänä tilaisuutta.</p></div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Mon, 08 Oct 2007 21:00:00 +0000 tkujala 370 at http://old.hiit.fi http://old.hiit.fi/node/347 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even">HIIT will organise on 9.10.2007 the NordicHIP seminar on themes on the Host Identity Protocol and the related internetworking themes. For more information, please contact Miika Komu. The seminar will be held in HIIT Spektri office.<br /> <br /> Seminar programme:<br /> <br /> Oleg Ponomarev: Bind/DNS experiments<br /> Antti Louko: HIP DNS Proxy + demo<br /> Samu Varjonen: HIP DHT Interface + demo?<br /> Miika Komu: NAT traversal<br /> Joakim Koskela: P2P SIP ovat HIP + demo<br /> Andrey Lukuanenko: Hi3 + demo<br /> Dmitry Korzun: Cyclic routing<br /> Laura Takkinen: HIP Privacy + demo<br /> Alberto Garcia: Opportunistic HIP + demo<br /> Bengt Ahlgren: Node ID Architecture<br /> Kristian Slavov: Secure migration to identity based networks<br /> Tobias Heer : PISA Wi-fi sharing model + demo<br /> Andrey Khurri/Samu Varjonen: Handover&nbsp; + demo</div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Mon, 08 Oct 2007 21:00:00 +0000 mam 347 at http://old.hiit.fi http://old.hiit.fi/node/372 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even">The HIIT-NRC seminar was held at Nokia Research Center. The objective of the event was to increase the transparency between HIIT and the Core Technology Centers of NRC. Some 20 persons were in attendance. [<a href="/node/371">Seminar programme and materials]</a></div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Tue, 02 Oct 2007 15:29:41 +0000 mam 372 at http://old.hiit.fi http://old.hiit.fi/node/356 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even">HIIT seminars in fall 2007 will be held in hall **B222** of Exactum,<br /> on Fridays starting at 10:15 a.m. Coffee available from 10.<br /> <br /> Sep 28:<br /> &nbsp; Samuel Kaski<br /> &nbsp; Relevance in Data Fusion and Visualization<br /> <br /> High-throughput measurement data, and data banks in which they are stored, have brought a new data analysis problem to biology and medicine: how to infer the relevant effects from the data mass. Each single information source, be it gene expression, protein interaction, or Gene Ontology, contains unknown amounts and types of noise, that is, irrelevant or uninteresting variation. The task of distinguishing between relevant and irrelevant variation is particularly hard in the initial exploratory task of &quot;looking at the data,&quot; when the hypotheses are still vague and hence there are no strong models yet to help constrain the exploration. I will discuss how to make sense of data masses with information visualization methods, and machine learning methods designed to bring out relevant clusters and components by fusing several data sources. Supervised mining or &quot;supervised unsupervised learning&quot; searches for clusters or components relevant or informative of classes such as gene ontology. Methods such as learning metrics, discriminative clustering, and discriminative components follow this principle. Mutual dependency mining separates task or source-specific variation from variation shared by all sources. The treatment-specific variation is less relevant in defining yeast stress response, for instance. Methods include local and non-parametric dependent components. Recently we have studied a related problem: How to use only partially relevant auxiliary data sources or background data in supporting inference on too few data samples.<br /> <br /> More information at http://www.cis.hut.fi/projects/mi</div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Thu, 27 Sep 2007 21:00:00 +0000 tkujala 356 at http://old.hiit.fi http://old.hiit.fi/node/362 <div class="field field-name-body field-type-text-with-summary field-label-hidden"><div class="field-items"><div class="field-item even">The launch event of the Finland-ICSI Center for Novel Internet Architectures (FICNIA) will be held in the <a href="http://www.icsi.berkeley.edu/">International Computer Science Institute</a> (ICSI) in Berkeley, California. <br /> <br /> The mission of the Center is to conduct fundamental research in novel Internet architectures, aiming at a significant contribution towards the future development of the Internet. The Center will combine and coordinate the research efforts at ICSI, led by Professor Scott Shenker, with those of HIIT and other Finnish reseachers, led by Professor Kimmo Raatikainen. It will operate through joint research activities, research visits, and events.<br /> <br /> The programme of the launch event is as follows:<br /> <br /> 09:30-09:45 &nbsp;&nbsp; Opening ceremonies<br /> <br /> &nbsp;&nbsp;&nbsp; Short notes: <br /> &nbsp;&nbsp;&nbsp; Professor Scott Shenker, ICSI, Co-director of the Center<br /> &nbsp;&nbsp;&nbsp; Professor Kimmo Raatikainen, HIIT, Co-director of the Center<br /> &nbsp;&nbsp;&nbsp; Professor Martti M&auml;ntyl&auml;, Director of HIIT<br /> &nbsp;&nbsp;&nbsp; Professor Nelson Morgan, Director of ICSI<br /> &nbsp;&nbsp;&nbsp; Mr. Riku M&auml;kel&auml;, TEKES<br /> <br /> 09:45-12:00&nbsp;&nbsp;&nbsp; Talks by HIIT visitors<br /> <br /> &nbsp;&nbsp;&nbsp; Professor Martti M&auml;ntyl&auml;: HIIT Overview<br /> &nbsp;&nbsp;&nbsp; Professor Kimmo Raatikainen: HIIT's Future Internet Programme<br /> &nbsp;&nbsp;&nbsp; Dr. Arto Karila: HIIT FI activities in EU 7th FP<br /> &nbsp;&nbsp;&nbsp; Dr. Sasu Tarkoma: Publish/Subscribe Internetworking<br /> &nbsp;&nbsp;&nbsp; Dr. Jukka Manner: UFOs on the Internet<br /> &nbsp;&nbsp;&nbsp; Dr. Kristiina Karvonen: Decision-making about trust in the Internet - the human perspective<br /> <br /> 12:00-01:00&nbsp;&nbsp;&nbsp; Break<br /> <br /> 01:00-02:30&nbsp;&nbsp;&nbsp; Talks by Berkeley students<br /> <br /> &nbsp;&nbsp; Byung-Gon Chun: Data-Oriented Network Architecture:<br /> &nbsp;&nbsp; Bowei Du: A Pub/Sub Communications API<br /> &nbsp;&nbsp; Daekyeong Moon: The Accountable Internet Protocol (AIP)<br /> &nbsp;&nbsp; Rodrigo Fonseca: X-Trace: A Pervasive Network Tracing Framework<br /> <br /> 03:00-04:30&nbsp;&nbsp;&nbsp; Planning session of the future activities of the Center</div></div></div><div class="field field-name-taxonomy-vocabulary-5 field-type-taxonomy-term-reference field-label-above"><div class="field-label">Events:&nbsp;</div><div class="field-items"><div class="field-item even"><a href="/events/2007">Events 2007</a></div></div></div> Wed, 26 Sep 2007 21:00:00 +0000 mam 362 at http://old.hiit.fi