Deep neural networks are increasingly used on mobile devices, where computational resources are limited. In this paper we develop CondenseNet, a novel network architecture with unprecedented efficiency. It combines dense connectivity with a novel module called learned group convolution. The dense connectivity facilitates feature re-use in the network, whereas learned group convolutions remove connections between layers for which this feature re-use is superfluous. At test time, our model can be implemented using standard group convolutions, allowing for efficient computation in practice. Our experiments show that Con-denseNets are far more efficient than state-of-the-art compact convolutional networks such as ShuffleNets.

Rendering bridges the gap between 2D vision and 3D scenes by simulating the physical process of image formation. By inverting such renderer, one can think of a learning approach to infer 3D information from 2D images. However, standard graphics renderers involve a fundamental discretization step called rasterization, which prevents the rendering process to be differentiable, hence able to be learned. Unlike the state-of-the-art differentiable renderers [29,19], which only approximate the rendering gradient in the back propagation, we propose a truly differentiable rendering framework that is able to (1) directly render colorized mesh using differentiable functions and (2) back-propagate efficient supervision signals to mesh vertices and their attributes from various forms of image representations, including silhouette, shading and color images.The key to our framework is a novel formulation that views rendering as an aggregation function that fuses the probabilistic contributions of all mesh triangles with respect to the rendered pixels. Such formulation enables our framework to flow gradients to the occluded and far-range vertices, which cannot be achieved by the previous state-of-thearts. We show that by using the proposed renderer, one can achieve significant improvement in 3D unsupervised singleview reconstruction both qualitatively and quantitatively. Experiments also demonstrate that our approach is able to handle the challenging tasks in image-based shape fitting, which remain nontrivial to existing differentiable renderers. Code is available at https://github.com/ ShichenLiu/SoftRas.

No abstract

Summary Some cancers originate from a single mutation event in a single cell. Blood cancers known as myeloproliferative neoplasms (MPNs) are thought to originate when a driver mutation is acquired by a hematopoietic stem cell (HSC). However, when the mutation first occurs in individuals and how it affects the behavior of HSCs in their native context is not known. Here we quantified the effect of the JAK2 -V617F mutation on the self-renewal and differentiation dynamics of HSCs in treatment-naive individuals with MPNs and reconstructed lineage histories of individual HSCs using somatic mutation patterns. We found that JAK2- V617F mutations occurred in a single HSC several decades before MPN diagnosis—at age 9 ± 2 years in a 34-year-old individual and at age 19 ± 3 years in a 63-year-old individual—and found that mutant HSCs have a selective advantage in both individuals. These results highlight the potential of harnessing somatic mutations to reconstruct cancer lineages.