scispace - formally typeset
Search or ask a question
Institution

Helsinki Institute for Information Technology

FacilityEspoo, Finland
About: Helsinki Institute for Information Technology is a facility organization based out in Espoo, Finland. It is known for research contribution in the topics: Population & Bayesian network. The organization has 630 authors who have published 1962 publications receiving 63426 citations.


Papers
More filters
Proceedings Article
27 Aug 2020
TL;DR: B batch-sequential Bayesian experimental design strategies to parallellise the expensive simulations and a numerical method to fully quantify the uncertainty in, for example, ABC posterior moments is proposed.
Abstract: The computational efficiency of approximate Bayesian computation (ABC) has been improved by using surrogate models such as Gaussian processes (GP). In one such promising framework the discrepancy between the simulated and observed data is modelled with a GP which is further used to form a model-based estimator for the intractable posterior. In this article we improve this approach in several ways. We develop batch-sequential Bayesian experimental design strategies to parallellise the expensive simulations. In earlier work only sequential strategies have been used. Current surrogate-based ABC methods also do not fully account the uncertainty due to the limited budget of simulations as they output only a point estimate of the ABC posterior. We propose a numerical method to fully quantify the uncertainty in, for example, ABC posterior moments. We also provide some new analysis on the GP modelling assumptions in the resulting improved framework called Bayesian ABC and discuss its connection to Bayesian quadrature (BQ) and Bayesian optimisation (BO). Experiments with toy and real-world simulation models demonstrate advantages of the proposed techniques.

12 citations

Proceedings Article
01 Jan 2019
TL;DR: In this paper, a Markov decision process is proposed to learn from a teacher that plans to recognize the teaching intent of the responses, and the learner is endowed with a model of the teacher.
Abstract: Machine teaching addresses the problem of finding the best training data that can guide a learning algorithm to a target model with minimal effort. In conventional settings, a teacher provides data that are consistent with the true data distribution. However, for sequential learners which actively choose their queries, such as multi-armed bandits and active learners, the teacher can only provide responses to the learner’s queries, not design the full data. In this setting, consistent teachers can be sub-optimal for finite horizons. We formulate this sequential teaching problem, which current techniques in machine teaching do not address, as a Markov decision process, with the dynamics nesting a model of the learner and the actions being the teacher's responses. Furthermore, we address the complementary problem of learning from a teacher that plans: to recognise the teaching intent of the responses, the learner is endowed with a model of the teacher. We test the formulation with multi-armed bandit learners in simulated experiments and a user study. The results show that learning is improved by (i) planning teaching and (ii) the learner having a model of the teacher. The approach gives tools to taking into account strategic (planning) behaviour of users of interactive intelligent systems, such as recommendation engines, by considering them as boundedly optimal teachers.

12 citations

Proceedings ArticleDOI
07 Sep 2010
TL;DR: The results of the intervention study in which a mobile hybrid media system that combines the use of the traditional printed book with the mobile phone was used in English as foreign language (EFL) education in primary school revealed an increase in learning motivation but also some conflicts when the boundaries of the school world and everyday life were blurred through theUse of new technology.
Abstract: The adoption of new technologies in primary schools has fallen behind in terms of children's everyday use of technology. The use of mobile phones has been proposed as a promising field for learning. To date, the mobile learning technologies have rarely been integrated with current educational practices, however. Here, we present the results of our intervention study in which a mobile hybrid media system that combines the use of the traditional printed book with the mobile phone was used in English as foreign language (EFL) education in primary school. The results revealed an increase in learning motivation but also some conflicts when the boundaries of the school world and everyday life were blurred through the use of new technology.

12 citations

Book ChapterDOI
21 Jun 2010
TL;DR: This work presents a novel and efficient algorithm for solving the most reliable subgraph problem with multiple query nodes on undirected random graphs, and demonstrates experimentally the usefulness of reliable k-terminal subgraphs.
Abstract: We present a novel and efficient algorithm for solving the most reliable subgraph problem with multiple query nodes on undirected random graphs Reliable subgraphs are useful for summarizing connectivity between given query nodes Formally, we are given a graph G=(V, E), a set of query (or terminal) nodes Q⊂V, and a positive integer B The objective is to find a subgraph H⊂G containing Q, such that H has at most B edges, and the probability that H is connected is maximized Previous algorithms for the problem are either computationally demanding, or restricted to only two query nodes Our algorithm extends a previous algorithm to handle k query nodes, where 2≤k≤|V| We demonstrate experimentally the usefulness of reliable k-terminal subgraphs, and the accuracy, efficiency and scalability of the proposed algorithm on real graphs derived from public biological databases.

12 citations


Authors

Showing all 632 results

NameH-indexPapersCitations
Dimitri P. Bertsekas9433285939
Olli Kallioniemi9035342021
Heikki Mannila7229526500
Jukka Corander6641117220
Jaakko Kangasjärvi6214617096
Aapo Hyvärinen6130144146
Samuel Kaski5852214180
Nadarajah Asokan5832711947
Aristides Gionis5829219300
Hannu Toivonen5619219316
Nicola Zamboni5312811397
Jorma Rissanen5215122720
Tero Aittokallio522718689
Juha Veijola5226119588
Juho Hamari5117616631
Network Information
Related Institutions (5)
Google
39.8K papers, 2.1M citations

93% related

Microsoft
86.9K papers, 4.1M citations

93% related

Carnegie Mellon University
104.3K papers, 5.9M citations

91% related

Facebook
10.9K papers, 570.1K citations

91% related

Performance
Metrics
No. of papers from the Institution in previous years
YearPapers
20231
20224
202185
202097
2019140
2018127