Generative Maximum Entropy Learning for Multiclass Classification

Maximum entropy approach to classification is very well studied in applied statistics and machine learning and almost all the methods that exists in literature are discriminative in nature. In this paper, we introduce a maximum entropy classification method with feature selection for large dimensional data such as text datasets that is generative in nature. To tackle the curse of dimensionality of large data sets, we employ conditional independence assumption (Naive Bayes) and we perform feature selection simultaneously, by enforcing a `maximum discrimination' between estimated class conditional densities. For two class problems, in the proposed method, we use Jeffreys (J) divergence to discriminate the class conditional densities. To extend our method to the multi-class case, we propose a completely new approach by considering a multi-distribution divergence: we replace Jeffreys divergence by Jensen-Shannon (JS) divergence to discriminate conditional densities of multiple classes. In order to reduce computational complexity, we employ a modified Jensen-Shannon divergence (JS(GM)), based on AM-GM inequality. We show that the resulting divergence is a natural generalization of Jeffreys divergence to a multiple distributions case. As far as the theoretical justifications are concerned we show that when one intends to select the best features in a generative maximum entropy approach, maximum discrimination using J-divergence emerges naturally in binary classification. Performance and comparative study of the proposed algorithms have been demonstrated on large dimensional text and gene expression datasets that show our methods scale up very well with large dimensional datasets.

Male Takeover and Infanticide in Nilgiri Langurs Semnopithecus johnii in the Western Ghats, India

Male takeover and infanticide are a widespread phenomenon among non-human primates, observed mostly in species with a relatively longer lactation in relation to gestation. In this study, we report for the first time an episode of male takeover and infanticide, and the rarely reported occurrence of an all-male band and female dispersal, in Nilgiri langurs, Semnopithecus johnii, in the Western Ghats, India. The new male was a member of an all-male band. After the takeover, the resident male and 3 juvenile males left the group and joined the all-male band. A female whose infant was killed was found missing after some days. There were significant changes in the patterns of social interactions among the resident group females soon after the male takeover, wherein the females spent less time on social interactions as compared to before and after the episode of takeover. The new male rarely interacted with the females soon after the takeover. We also observed that the resident group shifted its home range to a poorer quality habitat. (C) 2014 S. Karger AG, Basel

A literature survey on the real-time computational simulation of biological organs

This paper lists some references that could in some way be relevant in the context of the real-time computational simulation of biological organs, the research area being defined in a very broad sense. This paper contains 198 references.

Interplay of male traits, male mating strategies and female mate choice in the Asian elephant, Elephas maximus

Elaborate male traits with no apparent adaptive value may have evolved through female mate discrimination. Tusks are an elaborate male-only trait in the Asian elephant that could potentially influence female mate choice. We examined the effect of male body size, tusk possession and musth status on female mate choice in an Asian elephant population. Large/musth males received positive responses from oestrous females towards courtship significantly more often than did small/non-musth males. Young, tusked non-musth males attempted courtship significantly more often than their tuskless peers, and received more positive responses (though statistically insignificant) than did tuskless males. A positive response did not necessarily translate into mating because of mate-guarding by a dominant male. Female elephants appear to choose mates based primarily on traits such as musth that signal direct fertility benefits through increased sperm received than for traits such as tusks that may signal only indirect fitness benefits.