Helsinki ICT Research Events

This event feed aggregates content from the Research Events feeds from the Helsinki Institute for Information Technology HIIT, Aalto University Department of Computer Science, and the University of Helsinki Department of Computer Science.

  • The Crowd-Median Algorithm

    Dr. Antti Ukkonen, HIIT

    Abstract:

    The power of human computation is founded on the capabilities of humans to process qualitative information in a manner that is hard to reproduce with a computer. However, all machine learning algorithms rely on mathematical operations, such as sums, averages, least squares etc. that are less suitable for human computation.

    This paper is an effort to combine these two aspects of data processing. We consider the...

  • 17.03.2014 13:15–14:00
    HIIT seminar
    Aalto University, Computer Science Building, lecture hall T2

    Abstract:
    The power of human computation is founded on the capabilities of
    humans to process qualitative information in a manner that is hard to
    reproduce with a computer. However, all machine learning algorithms
    rely on mathematical operations, such as sums, averages, least squares
    etc. that are less suitable for human computation.
    ...

  • 24.02.2014 14:00–16:00
    HIIT seminar
    Open Innovation House, 2nd floor, sofa corner, Otaniementie 19B, Espoo
    Our monthly INUSE-seminar continues next Monday, 24.2.2014 14-16 in Otaniemi (more info below). The next presentation by Mikael Johnson concerns one of the significant under-researched topics of user involvement and user studies in design today:...
  • 21.02.2014 10:15–11:15
    HIIT seminar
    Exactum B119
    Title:
     
    MCMC-driven Adaptive Multiple Importance Sampling
     
    Abstract:
     
    Monte Carlo (MC) methods are widely used for statistical inference and stochastic optimization. A well-known class of MC methods is composed of importance sampling (IS) and its adaptive extensions (such as adaptive multiple IS and population MC). In this work...
  • On competitive recommendations

    Jara Uitto, ETH Zürich, Switzerland

    Abstract:

    We are given an unknown binary matrix, where the entries correspond to preferences of users on items. We want to find at least one 1-entry in each row with a minimum number of queries. The number of queries needed heavily depends on the input matrix and a straightforward competitive analysis yields bad results for any online algorithm. Therefore, we analyze our algorithm against a weaker offline algorithm...

  • 11.02.2014 13:15–14:00
    HIIT seminar
    Aalto University, Computer Science Building, lecture hall T5

    Abstract: We are given an unknown binary matrix, where the entries correspond to preferences of users on items. We want to find at least one 1-entry in each row with a minimum number of queries. The number of queries needed heavily depends on the input matrix and a straightforward competitive analysis yields bad results for any online algorithm. Therefore, we analyze our...

  • From Lovasz theta to kernel methods: A new connection between graph theory and machine learning

    Vinay Jethava, Chalmers University of Technology, Sweden

    Abstract:

    A number of machine learning problems involve analysis of graphs and have borrowed extensively from graph theory - spectral clustering, graph kernels, etc. In this talk, I?ll present our recent work on establishing a connection between Lovasz theta function, a powerful concept in graph theory which has been used heavily in...

  • 10.02.2014 13:15–14:00
    HIIT seminar
    Aalto University, Computer Science Building, lecture hall T2

    Abstract:

    A number of machine learning problems involve analysis of graphs and have borrowed extensively from graph theory - spectral clustering, graph kernels, etc. In this talk, I’ll present our recent work on establishing a connection between Lovasz theta function, a powerful concept in graph theory which has been used heavily in algorithms and...

  • 07.02.2014 10:15–11:15
    HIIT seminar
    Exactum B119
    Title:
     
    
Information Propagation in the Bitcoin Network
     
    Abstract:
     
    
Bitcoin is a digital currency that unlike traditional currencies does
not rely on a centralized authority. Instead Bitcoin relies on a
network of volunteers that collectively implement a replicated ledger
and verify transactions. In this paper we analyze how...
  • Collaborative Matrix Factorization for Predicting Drug-Target Interactions

    Prof. Hiroshi Mamitsuka, Kyoto University, Japan

    Abstract:

    Computationally predicting drug-target interactions is useful to discover potential new drugs (or targets). Currently, powerful machine learning approaches for this issue use not only known drug-target interactions but also drug and target similarities. Using similarities is well-accepted pharmacologically, since the two types of similarities...

Pages

Add to My Calendar