Matti Pöllä
Post-graduate Researcher, M.Sc. (Tech.)
Contact Information
| Visiting address: | Room B332 Department of Information and Computer Science Konemiehentie 2, Otaniemi campus, Espoo, Finland. |
| Mail address: |
Aalto University School of Science and Technology Department of Information and Computer Science PO Box 15400 FI-00076 Aalto Finland |
| Email: | matti.polla@tkk.fi |
| Tel: | +358-9-470 25115 |
| Fax: | +358-9-470 23277 |
Publications
- 18Matti Pöllä and Timo Honkela. Negative selection of written language using character multiset statistics. Journal of Computer Science and Technology, 25(6):1256–1266, November 2010. [More info] [See also: dx.doi.org ...]
- 17Matti Pöllä. A generative model for self/non-self discrimination in strings. In Proceedings of ICANNGA'09: International Conference on Adaptive and Natural Computing Algorithms, Lecture Notes in Computer Science. Springer-Verlag, April 2009. to appear. [More info]
- 16Matti Pöllä, Timo Honkela, and Teuvo Kohonen. Bibliography of self-organizing map (SOM) papers: 2002-2005 addendum. Technical report, Helsinki University of Technology, Department of Information and Computer Science, Espoo, Finland, 2009. Report TKK-ICS-R23. [More info]
- 15Matti Pöllä, Timo Honkela, and Xiao-Zhi Gao. Biologically inspired clustering: Comparing the neural and immune paradigms. In Proceedings of NICSO 2007 Workshop on Nature Inspired Cooperative Strategies for Optimization, volume 129 of Studies in Computational Intelligence (SCI), pages 179–188, Acireale, Italy, November 2008. Springer-Verlag. [More info]
- 14Matti Pöllä and Timo Honkela. Change detection of text documents using negative first-order statistics. In Poceedings of AKRR'08, The Second International and Interdisciplinary Conference on Adaptive Knowledge Representation and Reasoning, pages 48–55, Porvoo, Finland, September 2008. [More info]
- 13Mari-Sanna Paukkeri, Ilari Nieminen, Matti Pöllä, and Timo Honkela. A language-independent approach to keyphrase extraction and evaluation. In Proceedings of the 22nd International Conference on Computational Linguistics (COLING 2008), Manchester, UK, August 2008. to appear. [More info]
- 12Mats Sjöberg, Jorma Laaksonen, Timo Honkela, and Matti Pöllä. Inferring semantics from textual information in multimedia retrieval. Neurocomputing, 2008. to appear. [More info]
- 11Matti Pöllä and Timo Honkela. Probabilistic text change detection using an immune model. In Proceedings of the International Joint Conference on Neural Networks, IJCNN 2007, pages 1109–1114, Orlando, Florida, August 2007. [More info]
- 10Konstantinos Stamatakis, Konstantinos Chandrinos, Vangelis Karkaletsis, Miquel Angel Mayer, Dagmar Villarroel Gonzales, Martin Labský, Enrique Amigó, and Matti Pöllä. AQUA, a system assisting labelling experts assess health web resources. In Proceedings of 12th International Symposium for Health Information Management Research, iSHIMR 2007, Sheffield, UK, July 2007. to appear. [More info]
- 9Konstantinos Stamatakis, Vangelis Metsis, Vangelis Karkaletsis, Marek Raccent23uzi cka, Vojtech Svátek, Enrique Amigó Cabrera, and Matti Pöllä. Content collection for the labeling of health-related web content. In Proceedings of 11th Conference on Artificial Intelligence in Medicine (AIME 07), Amsterdam, The Netherlands, July 2007. to appear. [More info]
- 8Timo Honkela and Matti Pöllä. Describing rich content: Future directions for the semantic web. In New Developments in Artificial Intelligence and Semantic Web. Proceedings of the 12th Finnish Artificial Intelligence Conference STeP 2006, pages 143–148, Espoo, Finland, October 2006. Helsinki University of Technology. [More info]
- 7Matti Pöllä, Timo Honkela, Henrik Bruun, and Ann Russell. Analysis of interdisciplinary text corpora. In Proceedings of Ninth Scandinavian Conference on Artificial Intelligence (SCAI 2006), pages 17–22, Helsinki, Finland, October 2006. [More info]
- 6Mats Sjöberg, Jorma Laaksonen, Timo Honkela, and Matti Pöllä. Retrieval of multimedia objects by combining semantic information from visual and textual descriptors. In Proceedings of International Conference on Artificial Neural Networks, Athens, Greece, September 2006. 75–83. [More info] [See also: dx.doi.org ...]
- 5Matti Pöllä and Timo Honkela. Self-organizing neural network models for state anticipatory systems. In D. Dubois, editor, Computing Anticipatory Systems (CASYS'05), Liège, Belgium, August 2005. American Institute of Physics, Woodbury, New York. Received Best Paper Award. [More info]
- 4Matti Pöllä, Tiina Lindh-Knuutila, and Timo Honkela. Self-refreshing SOM as a semantic memory model. In Proceedings of AKRR'05, International and Interdisciplinary Conference on Adaptive Knowledge Representation and Reasoning, pages 171–174, Espoo, Finland, June 2005. [More info] [See also: www.cis.hut.fi ...]
- 3Matti Pöllä. Modeling anticipatory behavior with self-organizing neural networks. Master's thesis, Helsinki University of Technology, Espoo, Finland, May 2005. [More info] [See also: www.cis.hut.fi ...]
- 2Seppo Ovaska, Olli Vainio, and Matti Pöllä. Adaptive filtering using multiplicative general parameters for zero-crossing detection. IEEE Transactions on Industrial Electronics, 50(6):1340–1342, December 2003. [More info]
- 1Matti Pöllä. Comparison of predictive FIR-based zero-crossing detection methods. In Proceedings of FINSIG'03 – 2003 Finnish Signal Processing Symposium, pages 87–90, Tampere, Finland, May 2003. [More info] [See also: www.cis.hut.fi ...]
See my publications page (available also as [BibTeX] [ps] [pdf]). See also Google Scholar, DBLP.
Research Interests
Artificial immune systems in data mining
Artificial Immune Systems (AIS) are computational models developed with the vertebrate immune system as a motivation. My latest research involves applying AIS models for data mining of natural language data and developing new AIS algorithms for text mining.
Modeling Anticipatory Behavior with Self-Organizing Neural Networks
Cognitive tasks performed by humans are often driven by anticipations about the future. Traditionally, the AI of an agent is implemented as a reactive if--then rule set, which allows the agent to behave only reactively. An internal predictive model can assist an agent to simulate future events and thus act anticipatorily. In my research the focus has been on building a prototype-based neural network estimate of the dynamic state space of the agent.
Sequential Learning and Forgetting in the Self Organizing Map
The Self-Organizing Map (SOM) is a visualization and clustering tool for creating a topologically correct mapping of a high-dimensional data set into a two-dimensional neuron lattice. The SOM is typically trained with a static set of data and thus all inputs are equally represented in the SOM projection. However, when training the SOM sequentially, representations of old inputs may overwritten, which can be understood as a manifestation of catastrophic forgetting found in feed-forward networks. Self-refreshing mechanism based on generating pseudo-data using the SOM codebook vectors can make the forgetting process a gradual instead of catastrophic.
Language Learning in Multi-Agent Systems
A group of autonomous agents operating in the same environment can benefit from inter-agent communication. The emergence of an shared language is based on finding a common semantic association between percept objects and utterances describing their properties. In the SOMAgent framework the semantic memory association task is implemented using the SOM.
Self-Organizing Map -demos
I have written some Java applets to demonstrate the Self-Organizing Map algorithm.
Teaching
- T-61.3030 Principles of Neural Computing (5 cr )
- T-61.5130 Machine Learning and Neural Networks (5 cr )
Links
I also have a personal home page.