Skin lesions on blue whales off southern Chile: Possible conservation implications?

We report on three types of skin lesions in a population of blue whales, Balaenoptera musculus, off the northwestern coast of Isla Grande de Chiloe, Chile. These lesions were: (1) cookie-cutter shark, Isistius brasilensis, bites, (2) vesicular or blister lesions, and (3) a tattoo-like skin disease. The presence of these lesions was determined by the examining photos collected in 2006 and 2007 for a blue whale photo-identification project. We examined 289 photographs of 68 individuals for lesions. The cookie-cutter shark lesions are common on these blue whales and similar to those reported from other species of cetaceans. Skin peeling or shedding was observed on some whales and is believed to be a normal condition. Based on the photographs examined to date the vesicular lesions are more common than the tattoo-like lesions. The tattoo-like skin lesions was observed just on a single whale in 2007. The blister lesions were common on whales in both 2006 and 2007. The presence of blister lesions in both years may indicate that this “disease” will be present in the population for a long time. It is unknown if these lesions contribute to mortality of blue whales frequenting Chilean waters, but the tattoo-like skin lesions if shown to be a pox virus could cause neonatal and calf mortality. Additional investigations are needed that, as a minimum, must include the histological and genetic examination of the two types of disease from live or dead whales, especially the tattoo-like skin lesions. Until this work is undertaken, it will be impossible to determine if these lesions pose a conservation risk to the blue whales off Chile.

Bandit-Based Algorithms for Budgeted Learning

We explore the problem of budgeted machine learning, in which the learning algorithm has free access to the training examples’ labels but has to pay for each attribute that is specified. This learning model is appropriate in many areas, including medical applications. We present new algorithms for choosing which attributes to purchase of which examples in the budgeted learning model based on algorithms for the multi-armed bandit problem. All of our approaches outperformed the current state of the art. Furthermore, we present a new means for selecting an example to purchase after the attribute is selected, instead of selecting an example uniformly at random, which is typically done. Our new example selection method improved performance of all the algorithms we tested, both ours and those in the literature.

Splinter Dedication

This man who grew up on an irrigated farm near North Platte was inducted into the National Academy of Engineering in 1984 for invention and development of aerial-spray systems and harvesting systems, the only Nebraskan elected to that esteemed academy. Six patents are based on his research.