Gut microbes react to a meal — but have no use for nutrition labels | Cell Host Microbe
The influence of diet on gut microbes seems to be highly personal, according to a comprehensive analysis of microbiome complexity. The results suggest that the type of food consumed — for example, a leafy vegetable versus a piece of meat — is a better indicator of changes to the microbiome than is the food’s nutrition label. The team also found that a monotonous diet did not correspond to a stable microbiome: two participants who relied primarily on the meal-replacement beverage Soylent for nourishment still showed changes in their microbiomes over time.
Deep Neural Networks Improve Radiologists’ Performance in Breast Cancer Screening | arXiv
Researchers present a deep convolutional neural network for breast cancer screening exam classification, trained and evaluated on over 200,000 exams (over 1,000,000 images). Their network achieves an AUC of 0.895 in predicting whether there is a cancer in the breast, when tested on the screening population.
Weight Agnostic Neural Networks | arXiv
Not all neural network architectures are created equal, some perform much better than others for certain tasks. But how important are the weight parameters of a neural network compared to its architecture? In this work, researchers question to what extent neural network architectures alone, without learning any weight parameters, can encode solutions for a given task. They propose a search method for neural network architectures that can already perform a task without any explicit weight training.
Research paper | Interactive article