Semantic Discriminant mapping for classification and browsing of remote sensing textures and objects

We present a new approach based on Discriminant Analysis to map a high dimensional image feature space onto a subspace which has the following advantages: 1. each dimension corresponds to a semantic likelihood, 2. an efficient and simple multiclass classifier is proposed and 3. it is low dimensional. This mapping is learnt from a given set of labeled images with a class groundtruth. In the new space a classifier is naturally derived which performs as well as a linear SVM. We will show that projecting images in this new space provides a database browsing tool which is meaningful to the user. Results are presented on a remote sensing database with eight classes, made available online. The output semantic space is a low dimensional feature space which opens perspectives for other recognition tasks. © 2005 IEEE.

Extending noise robust structured support vector machines to larger vocabulary tasks

This paper describes a structured SVM framework suitable for noise-robust medium/large vocabulary speech recognition. Several theoretical and practical extensions to previous work on small vocabulary tasks are detailed. The joint feature space based on word models is extended to allow context-dependent triphone models to be used. By interpreting the structured SVM as a large margin log-linear model, illustrates that there is an implicit assumption that the prior of the discriminative parameter is a zero mean Gaussian. However, depending on the definition of likelihood feature space, a non-zero prior may be more appropriate. A general Gaussian prior is incorporated into the large margin training criterion in a form that allows the cutting plan algorithm to be directly applied. To further speed up the training process, 1-slack algorithm, caching competing hypothesis and parallelization strategies are also proposed. The performance of structured SVMs is evaluated on noise corrupted medium vocabulary speech recognition task: AURORA 4. © 2011 IEEE.

Increasing the open-circuit voltage of photoprotein-based photoelectrochemical cells by manipulation of the vacuum potential of the electrolytes.

The innately highly efficient light-powered separation of charge that underpins natural photosynthesis can be exploited for applications in photoelectrochemistry by coupling nanoscale protein photoreaction centers to man-made electrodes. Planar photoelectrochemical cells employing purple bacterial reaction centers have been constructed that produce a direct current under continuous illumination and an alternating current in response to discontinuous illumination. The present work explored the basis of the open-circuit voltage (V(OC)) produced by such cells with reaction center/antenna (RC-LH1) proteins as the photovoltaic component. It was established that an up to ~30-fold increase in V(OC) could be achieved by simple manipulation of the electrolyte connecting the protein to the counter electrode, with an approximately linear relationship being observed between the vacuum potential of the electrolyte and the resulting V(OC). We conclude that the V(OC) of such a cell is dependent on the potential difference between the electrolyte and the photo-oxidized bacteriochlorophylls in the reaction center. The steady-state short-circuit current (J(SC)) obtained under continuous illumination also varied with different electrolytes by a factor of ~6-fold. The findings demonstrate a simple way to boost the voltage output of such protein-based cells into the hundreds of millivolts range typical of dye-sensitized and polymer-blend solar cells, while maintaining or improving the J(SC). Possible strategies for further increasing the V(OC) of such protein-based photoelectrochemical cells through protein engineering are discussed.

Discriminative spoken language understanding using word confusion networks

Current commercial dialogue systems typically use hand-crafted grammars for Spoken Language Understanding (SLU) operating on the top one or two hypotheses output by the speech recogniser. These systems are expensive to develop and they suffer from significant degradation in performance when faced with recognition errors. This paper presents a robust method for SLU based on features extracted from the full posterior distribution of recognition hypotheses encoded in the form of word confusion networks. Following [1], the system uses SVM classifiers operating on n-gram features, trained on unaligned input/output pairs. Performance is evaluated on both an off-line corpus and on-line in a live user trial. It is shown that a statistical discriminative approach to SLU operating on the full posterior ASR output distribution can substantially improve performance both in terms of accuracy and overall dialogue reward. Furthermore, additional gains can be obtained by incorporating features from the previous system output. © 2012 IEEE.