Struktura a její využití při rozpoznávání

Publikace

Autori Název
P. Černo Clearing Restarting Automata and Grammatical Inference
Proceedings of the Eleventh International Conference on Grammatical Inference (ICGI 2012), September 5-8, 2012, University of Maryland, College Park, United States, JMLR, Vol. 21, 54-68.
Abstrakt: Clearing and subword-clearing restarting automata are linguistically motivated models of automata. We investigate the problem of grammatical inference for such automata based on the given set of positive and negative samples. We show that it is possible to identify these models in the limit. In this way we can learn a large class of languages. On the other hand, we prove that the task of finding a clearing restarting automaton consistent with a given set of positive and negative samples is NP-hard, provided that we impose an upper bound on the width of its instructions.
Odkazy: BibTeX, JMLR Proceedings, Fulltext, Presentation
P. Černo Clearing Restarting Automata and Grammatical Inference (Technical Report)
Technical report, 1/2012, Charles University, Faculty of Mathematics and Physics, Prague.
Abstrakt: See above
Odkazy: BibTeX, Fulltext
I. Mrázová, M. Kukačka Can Deep Neural Networks Discover Meaningful Pattern Features?
In Procedia Computer Science, Vol. 12, pp. 194-199, 2012.
Abstrakt: Recent advances in the area of deep neural networks brought a lot of attention to some of the key issues important for their design. In particular for 2D-shapes, their accuracy has been shown to outperform all other classifiers - e.g., in the German Traffic Sign Competition run by IJCNN 2011. On the other hand, their training may be quite cumbersome and the structure of the network has to be chosen beforehand. This paper introduces a new sensitivity-based approach capable of picking the right image features from a pre-trained SOM-like feature detector. Experimental results obtained so far for hand-written digit recognition show that pruned network architectures impact a transparent representation of the features actually present in the data while improving network robustness.
Odkazy: BibTeX, ScienceDirect
I. Mrázová, M. Kukačka Can Deep Convolutional Neural Networks Discover Meaningful Pattern Features?
Technical report No 2012/1/KTIML, MFF UK, 12 p., 2012.
Abstrakt: Recent advances in the area of deep neural networks brought a lot of attention to some of the key issues important for their design. In particular for 2D-shapes, their accuracy has been shown to outperform all other classifiers - e.g., in the German Traffic Sign competition run by IJCNN 2011. While the majority of classical image processing techniques is based on carefully pre-selected image features, convolutional neural networks are, namely, designed to learn local features autonomously. On the other hand, the entire training process may be quite cumbersome and the structure of the network has to be chosen beforehand. Our paper introduces a new sensitivity-based approach capable of picking the right image features from a pretrained SOM-like feature detector. Experimental results obtained so far for two case studies on face and hand-written digit recognition show that pruned network architectures impact a transparent representation of the features actually present in the data while improving network robustness.
Odkazy: BibTeX
F. Otto, P. Černo, F. Mráz Limited Context Restarting Automata and McNaughton Families of Languages
In Rudolf Freund, Markus Holzer, Bianca Truthe, and Ulrich Ultes-Nitsche, editors, Workshop on Non-Classical Models of Automata and Applications (NCMA), books@ocg.at, pages 165-180. Österreichisches Computer Gesellschaft, 2012.
Abstrakt: In the literature various types of restarting automata have been studied that are based on contextual rewriting. A word w is accepted by such an automaton if, starting from the initial configuration that corresponds to input w, the word w is reduced to the empty word within a finite number of applications of these contextual rewritings. This approach is reminiscent of the notion of McNaughton families of languages. Here we put the aforementioned types of restarting automata into the context of McNaughton families of languages, relating the classes of languages accepted by these automata in particular to the class GCSL of growing context-sensitive languages and to the class CRL of Church-Rosser languages.
Odkazy: BibTeX, NCMA 2012, Presentation
J. Pihera Artificial neural networks and their application for 3D-data processing
MSc-thesis (supervised by I. Mrazova), KTIML MFF UK, 135 p., 2012. This thesis won Dean's Award for the Best MSc-thesis in the Field of Computer Science.
Abstrakt: Neural networks represent a powerful means capable of processing various multi-media data. Two applications of artificial neural networks to 3D surface models are examined in this thesis – detection of significant features in 3D data and model classification. The theoretical review of existing self-organizing neural networks is presented and followed by description of feed-forward neural networks and convolutional neural networks (CNN). A novel modification of the existing model – N-dimensional convolutional neural networks (ND-CNN) – is introduced. The proposed ND-CNN model is enhanced by an existing technique for enforced knowledge representation. The developed theoretical methods are assessed on supporting experiments with scanned 3D face models. The first experiment focuses on automatic detection of significant facial features while the second experiment performs classification of the models by their gender using the CNN and ND-CNN.
Odkazy: Dean's Award

Poslední úprava: Neděle, 9.3.2014