Scaling the Indian Buffet Process via Submodular Maximization

Inference for latent feature models is inherently difficult as the inference space grows exponentially with the size of the input data and number of latent features. In this work, we use Kurihara & Welling (2008)'s maximization-expectation framework to perform approximate MAP inference for linear-Gaussian latent feature models with an Indian Buffet Process (IBP) prior. This formulation yields a submodular function of the features that corresponds to a lower bound on the model evidence. By adding a constant to this function, we obtain a nonnegative submodular function that can be maximized via a greedy algorithm that obtains at least a one-third approximation to the optimal solution. Our inference method scales linearly with the size of the input data, and we show the efficacy of our method on the largest datasets currently analyzed using an IBP model.

Influence of selected Indian immunostimulant herbs against white spot syndrome virus (WSSV) infection in black tiger shrimp, Penaeus monodon with reference to haematological, biochemical and immunological changes

Immunostimulants are the substances, which enhance the non-specific defence mechanism and provide resistance against the invading pathogenic micro-organism. In order to increase the immunity of shrimps against the WSSV, the methanolic extracts of five different herbal medicinal plants like Cyanodon dactylon, Aegle marmelos, Tinospora cordifolia, Picrorhiza kurooa and Eclipta alba were selected and mixed thoroughly in equal proportion. The mixed extract was supplemented with various concentrations viz. 100 (A), 200 (B), 400 (C), and 800 (D) mg kg(-1) through artificial diets individually. The prepared diets (A-D) were fed individually to WSSV free healthy shrimp Penaeus monodon with an average weight of 8.0 +/- 0.5 g for 25 days. Control diet (E), devoid of herbal extract was also fed to shrimps simultaneously. After 25 days of feeding experiment, the shrimps were challenged with WSSV, which were isolated and propagated from the infected crustaceans. The shrimps succumbed to death within 7 days when fed on no herbal immunostimulant diet (E). Among the different concentrations of herbal immunostimulant supplemented diets, the shrimps fed on diet D (800 mg kg(-1)) significantly (P < 0.0001) had more survival (74%) and reduction in the viral load. Also the better performance of haematological, biochemical and immunological parameters was found in the immunostimulant incorporated diets fed shrimps. The present work revealed that the application of herbal immunostimulants will be effective against shrimp viral pathogenesis and they can be recommended for shrimp culture. (c) 2006 Published by Elsevier Ltd.