Refine
Year of publication
- 2006 (60) (remove)
Document Type
- Article (38)
- Monograph/Edited Volume (12)
- Doctoral Thesis (9)
- Master's Thesis (1)
Is part of the Bibliography
- yes (60) (remove)
Keywords
- EEG (2)
- Maschinelles Lernen (2)
- 3D-Stadtmodelle (1)
- 3d city models (1)
- Ackerschmalwand (1)
- Aufzählung (1)
- BCI (1)
- Biocomputing (1)
- Bioinformatik (1)
- Brain Computer Interface (1)
- Classification (1)
- Code (1)
- Codierung (1)
- Common Spatial Pattern (1)
- Computergrafik (1)
- Computersicherheit (1)
- DNA (1)
- DNA computing (1)
- DNS (1)
- Dynamische Rekonfiguration (1)
- E-Learning (1)
- Feature Combination (1)
- Feedback (1)
- Fehlende Daten (1)
- Gehirn-Computer-Schnittstelle (1)
- Geovisualisierung (1)
- Hauptkomponentenanalyse (1)
- IT security (1)
- Information Transfer Rate (1)
- Kybernetik (1)
- LBA problem (1)
- Lindenmayer systems (1)
- Machine Learning (1)
- Middleware (1)
- Multi-Class (1)
- Neuronales Netz (1)
- Signal Processing (1)
- Single Trial Analysis (1)
- Spatio-Spectral Filter (1)
- Suche (1)
- Temporäre Anbindung (1)
- Texturen (1)
- Unabhängige Komponentenanalyse (1)
- VM (1)
- Web Services (1)
- accepting grammars (1)
- bio-computing (1)
- brain-computer interface (1)
- code (1)
- common spatial patterns (1)
- computer graphics (1)
- degree of non-regulation (1)
- drug discovery (1)
- dynamic reconfiguration (1)
- enumeration (1)
- event-related desynchronization (1)
- finite state sequential transducers (1)
- geovisualization (1)
- incremental SVM (1)
- intrusion detection (1)
- leftmost derivations (1)
- molecular networks (1)
- molekulare Netzwerke (1)
- nichtlineare PCA (NLPCA) (1)
- nonlinear PCA (NLPCA) (1)
- online learning (1)
- programmed grammars (1)
- search (1)
- single-trial-analysis (1)
- state complexity (1)
- temporary binding (1)
- textures (1)
- virtual machine (1)
Institute
- Institut für Informatik und Computational Science (60) (remove)
Incremental Support Vector Machines (SVM) are instrumental in practical applications of online learning. This work focuses on the design and analysis of efficient incremental SVM learning, with the aim of providing a fast, numerically stable and robust implementation. A detailed analysis of convergence and of algorithmic complexity of incremental SVM learning is carried out. Based on this analysis, a new design of storage and numerical operations is proposed, which speeds up the training of an incremental SVM by a factor of 5 to 20. The performance of the new algorithm is demonstrated in two scenarios: learning with limited resources and active learning. Various applications of the algorithm, such as in drug discovery, online monitoring of industrial devices and and surveillance of network traffic, can be foreseen.
Background: The biological interpretation of large-scale gene expression data is one of the paramount challenges in current bioinformatics. In particular, placing the results in the context of other available functional genomics data, such as existing bio-ontologies, has already provided substantial improvement for detecting and categorizing genes of interest. One common approach is to look for functional annotations that are significantly enriched within a group or cluster of genes, as compared to a reference group. Results: In this work, we suggest the information-theoretic concept of mutual information to investigate the relationship between groups of genes, as given by data-driven clustering, and their respective functional categories. Drawing upon related approaches (Gibbons and Roth, Genome Research 12: 1574-1581, 2002), we seek to quantify to what extent individual attributes are sufficient to characterize a given group or cluster of genes. Conclusion: We show that the mutual information provides a systematic framework to assess the relationship between groups or clusters of genes and their functional annotations in a quantitative way. Within this framework, the mutual information allows us to address and incorporate several important issues, such as the interdependence of functional annotations and combinatorial combinations of attributes. It thus supplements and extends the conventional search for overrepresented attributes within a group or cluster of genes. In particular taking combinations of attributes into account, the mutual information opens the way to uncover specific functional descriptions of a group of genes or clustering result. All datasets and functional annotations used in this study are publicly available. All scripts used in the analysis are provided as additional files.
Iterated finite state sequential transducers are considered as language generating devices. The hierarchy induced by the size of the state alphabet is proved to collapse to the fourth level. The corresponding language families are related to the families of languages generated by Lindenmayer systems and Chomsky grammars. Finally, some results on deterministic and extended iterated finite state transducers are established.
Non-stationarities are ubiquitous in EEG signals. They are especially apparent in the use of EEG-based brain- computer interfaces (BCIs): (a) in the differences between the initial calibration measurement and the online operation of a BCI, or (b) caused by changes in the subject's brain processes during an experiment (e.g. due to fatigue, change of task involvement, etc). In this paper, we quantify for the first time such systematic evidence of statistical differences in data recorded during offline and online sessions. Furthermore, we propose novel techniques of investigating and visualizing data distributions, which are particularly useful for the analysis of (non-) stationarities. Our study shows that the brain signals used for control can change substantially from the offline calibration sessions to online control, and also within a single session. In addition to this general characterization of the signals, we propose several adaptive classification schemes and study their performance on data recorded during online experiments. An encouraging result of our study is that surprisingly simple adaptive methods in combination with an offline feature selection scheme can significantly increase BCI performance
The Berlin Brain-Computer Interface (BBCI) project develops a noninvasive BCI system whose key features are 1) the use of well-established motor competences as control paradigms, 2) high-dimensional features from 128-channel electroencephalogram (EEG), and 3) advanced machine learning techniques. As reported earlier, our experiments demonstrate that very high information transfer rates can be achieved using the readiness potential (RP) when predicting the laterality of upcoming left-versus right-hand movements in healthy subjects. A more recent study showed that the RP similarily accompanies phantom movements in arm amputees, but the signal strength decreases with longer loss of the limb. In a complementary approach, oscillatory features are used to discriminate imagined movements (left hand versus right hand versus foot). In a recent feedback study with six healthy subjects with no or very little experience with BCI control, three subjects achieved an information transfer rate above 35 bits per minute (bpm), and further two subjects above 24 and 15 bpm, while one subject could not achieve any BCI control. These results are encouraging for an EEG-based BCI system in untrained subjects that is independent of peripheral nervous system activity and does not rely on evoked potentials even when compared to results with very well-trained subjects operating other BCI systems