We propose a deep convolutional neural network architecture codenamed Inception that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. By a carefully crafted design, we increased the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the Hebbian principle and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC14 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.

/pdf/going-deeper-with-convolutions-1yobw2o2ds.pdf

Going deeper with convolutions

There is, I think, something ethereal about i —the square root of minus one. I remember first hearing about it at school. It seemed an odd beast at that time—an intruder hovering on the edge of reality.

Usually familiarity dulls this sense of the bizarre, but in the case of i it was the reverse: over the years the sense of its surreal nature intensified. It seemed that it was impossible to write mathematics that described the real world in …

I and i

다중혈관 관상동맥 환자에서 y-문합을 이용하여 양쪽 내흉동맥만을 사용한 우회술의 조기 성적

抗原变异可使得多种致病微生物易于逃避宿主免疫应答。表达在感染红细胞表面的恶性疟原虫红细胞表面蛋白1（PfPMP1）与感染红细胞、内皮细胞、树突状细胞以及胎盘的单个或多个受体作用，在黏附及免疫逃避中起关键的作用。每个单倍体基因组var基因家族编码约60种成员，通过启动转录不同的var基因变异体为抗原变异提供了分子基础。

疟原虫var基因转换速率变化导致抗原变异[英]／Paul H, Robert P, Christodoulou Z, et al//Proc Natl Acad Sci U S A

Image-to-image translation is a class of vision and graphics problems where the goal is to learn the mapping between an input image and an output image using a training set of aligned image pairs. However, for many tasks, paired training data will not be available. We present an approach for learning to translate an image from a source domain X to a target domain Y in the absence of paired examples. Our goal is to learn a mapping G : X → Y such that the distribution of images from G(X) is indistinguishable from the distribution Y using an adversarial loss. Because this mapping is highly under-constrained, we couple it with an inverse mapping F : Y → X and introduce a cycle consistency loss to push F(G(X)) ≈ X (and vice versa). Qualitative results are presented on several tasks where paired training data does not exist, including collection style transfer, object transfiguration, season transfer, photo enhancement, etc. Quantitative comparisons against several prior methods demonstrate the superiority of our approach.

Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks

Arrangements of curves in the plane are of fundamental significance in many problems of computational and combinatorial geometry (e.g. motion planning, algebraic cell decomposition, etc.). In this paper we study various topological and combinatorial properties of such arrangements under some mild assumptions on the shape of the curves, and develop basic tools for the construction, manipulation, and analysis of these arrangements. Our main results include a generalization of the zone theorem of [EOS], [CGL] to arrangements of curves (in which we show that the combinatorial complexity of the zone of a curve is nearly linear in the number of curves), and an application of (some weaker variant of) that theorem to obtain a nearly quadratic incremental algorithm for the construction of such arrangements.

Arrangements of Curves in the Plane- Topology, Combinatorics, and Algorithms

Consider a collection of disjoint polygons in the plane containing a total of n edges. We show how to build, in O(n2) time and space, a data structure from which in O(n) time we can compute the visibility polygon of a given point with respect to the polygon collection. As an application of this structure, the visibility graph of the given polygons can be constructed in O(n2) time and space. This implies that the shortest path that connects two points in the plane and avoids the polygons in our collection can be computed in O(n2) time, improving earlier O(n2 log n) results.

Visibility-polygon search and euclidean shortest paths

This paper presents a scalable distributed algorithm for computing and maintaining multi-target identity information. The algorithm builds on a novel representational framework, Identity-Mass Flow, to overcome the problem of exponential computational complexity in managing multi-target identity explicitly. The algorithm uses local information to efficiently update the global multi-target identity information represented as a doubly stochastic matrix, and can be efficiently mapped to nodes in a wireless ad hoc sensor network. The paper describes a distributed implementation of the algorithm in sensor networks. Simulation results have validated the Identity-Mass Flow framework and demonstrated the feasibility of the algorithm.

/pdf/a-distributed-algorithm-for-managing-multi-target-identities-1pmjn3amb2.pdf

A distributed algorithm for managing multi-target identities in wireless ad-hoc sensor networks

\emph{We propose a new randomized algorithm for maintaining a set of c lusters among moving nodes in the plane Given a specified cluster radius, our algorithm selects and maintains a variable subset of the nodes as cluster centers This subset has the property that (1) balls of the given radius centered at the chosen nodes cover all the others and (2) the number of centers selected is a constant-factor approximation of the minimum possible As the nodes move, an event-based kinetic data structure updates the clustering as necessary This kinetic data structure is shown to be responsive, efficient, local, and compact The produced cover is also smooth, in the sense that wholesale cluster re-arrangements are avoided The algorithm can be implemented without exact knowledge of the node positions, if each node is able to sense its distance to other nodes up to the cluster radius Such a kinetic clustering can be used in numerous applications where mobile devices must be interconnected into an ad-hoc network to collaboratively perform some task}

/pdf/discrete-mobile-centers-4l7lnjflb3.pdf

Discrete mobile centers

We show that, under reasonable assumptions, any collision-avoiding motion planning problem for a moving system with two degrees of freedom can be solved in time O(ls(n) log2n), where n is the number of collision constraints imposed on the system, s is a fixed parameter depending e.g. on the maximum algebraic degree of these constraints, and ls(n) is the (almost linear) maximum length of (n,s) Davenport Schinzel sequences. This follows from an upper bound of O(ls(n)) that we establish for the combinatorial complexity of a single connected component of the space of all free placements of the moving system. Although our study is motivated by motion planning, it is actually a study of topological, combinatorial, and algorithmic issues involving a single face in an arrangement of curves. Our results thus extend beyond the area of motion planning, and have applications in many other areas.

/pdf/on-the-general-motion-planning-problem-with-two-degrees-of-4ehhq36bjt.pdf

Leonidas J. Guibas

Papers

Arrangements of Curves in the Plane- Topology, Combinatorics, and Algorithms

Visibility-polygon search and euclidean shortest paths

A distributed algorithm for managing multi-target identities in wireless ad-hoc sensor networks

Discrete mobile centers

On the general motion planning problem with two degrees of freedom