This paper presents a machine learning system for parsing natural language that learns from manually parsed example sentences, and parses unseen data at state-of-the-art accuracies. Its machine learning technology, based on the maximum entropy framework, is highly reusable and not specific

examples of animal–bacterial associations have arisen as a result. Thus far, it has been difficult for biologists to design ways of determining the extent to which these associations have influenced the biology of animals, including their developmental patterns. The following review focuses on an emerging

Abstract Three case studies failed to demonstrate impacts of early environment or maternal effects on breeding in situations where they could have been expected. This leads to a number of methodological questions about the resolving power required to detect such impacts, but above all else to the conclusion that maternal effects and homeotic control are opposites. When assessing potential maternal effects, one has to consider not only the developmental period in which they occur but also the later stage of life of concern, because, with age, maternal effects may become less and less important or disappear altogether. The only real measure that there is of the relative importance of early environment and maternal effects is their proportion in phenotypic variance in the traits of interest.

of t. We propose an alternative estimator which possesses the essential properties, and is easily obtained. We illustrate our estimator with two numerical examples and a simulation study.

The omnipresence of unknown words is a problem that any NLP component needs to address in some form. While there exist many established techniques for dealing with unknown words in the realm of POS-tagging, for example, guessing unknown words ’ semantic properties is a less-explored area

We study the problem of how to recognize a new human ac-tivity when we have never seen any training example of that activity before. Recognizing human activities is an essen-tial element for user-centric and context-aware applications. Previous studies showed promising results using various ma

with a stereo-head mounted on an autonomous ground vehicle. We give examples of camera trajectories estimated in real-time purely from images over previously unseen distances (600 meters) and periods of time. 1.

the joint space of attribute ranking outputs, and propose a novel form of zero-shot learning in which the supervisor relates the unseen object category to previously seen objects via attributes (for example, ‘bears are furrier than giraffes’). We further show how the proposed relative attributes enable

of training examples such that the chosen hypothesis is predicted to characterize unseen examples...

In multi-label learning, each training example is associated with a set of labels and the task is to predict the proper label set for the unseen example. Due to the tremendous (expo-nential) number of possible label sets, the task of learning from multi-label examples is rather challenging