Recursive partition and amalgamation with the exponential family: Theory and applications

The theory of tree-growing (RECPAM approach) is developed for outcome variables which are distributed as the canonical exponential family. The general RECPAM approach (consisting of three steps: recursive partition, pruning and amalgamation), is reviewed. This is seen as constructing a partition with maximal information content about a parameter to be predicted, followed by simplification by the elimination of ‘negligible’ information. The measure of information is defined for an exponential family outcome as a deviance difference, and appropriate modifications of pruning and amalgamation rules are discussed. It is further shown how the proposed approach makes it possible to develop tree-growing for situations usually treated by generalized linear models (GLIM). In particular, Poisson and logistic regression can be tree-structured. Moreover, censored survival data can be treated, as in GLIM, by observing a formal equivalence of the likelihood under random censoring and an appropriate Poisson model. Three examples are given of application to Poisson, binary and censored survival data.     

Cyclic permutations and evolutionary trees

Given a tree with leaf set X, there are certain ways of arranging the elements of X in a circular order so that can be embedded in the plane and ‘preserve’ this ordering. We investigate some new combinatorial properties of these ‘circular orderings.’ We then use these properties to establish two results concerning dissimilarity maps on X that are induced by edge-weighted trees with leaf set X.     

因素空间理论与知识表示的数学框架(XⅢ)──描述架中概念的结构(2)

本文是［１，１２］的继续，研究描述架中概念的结构；本文讨论后半部分，内容涉及概念内涵与外延的转换，清晰关系的内投影与内变换，概念的结构＊以及有关问题的注记。     

树映射的分布混沌

设T是个有限树,f是T上的连续映射．证明了f是分布混沌的当且仅当它的拓扑熵是正数．一些已知结论得到了改进．     

Decomposition algorithms for the tree edit distance problem

We study the behavior of dynamic programming methods for the tree edit distance problem, such as [P. Klein, Computing the edit-distance between unrooted ordered trees, in: Proceedings of 6th European Symposium on Algorithms, 1998, p. 91–102; K. Zhang, D. Shasha, SIAM J. Comput. 18 (6) (1989) 1245–1262]. We show that those two algorithms may be described as decomposition strategies. We introduce the general framework of cover strategies, and we provide an exact characterization of the complexity of cover strategies. This analysis allows us to define a new tree edit distance algorithm, that is optimal for cover strategies.     

光链路设计的Java小程序实现

介绍光链路设计的原理和公式 ,给出光链路设计的Java小程序实现 ,以期有效解决在手工进行光链路设计及计算中存在的问题。     

Deep ensemble-based hard sample mining for food recognition

Deep neural networks represent a compelling technique to tackle complex real-world problems, but are over-parameterized and often suffer from over- or under-confident estimates. Deep ensembles have shown better parameter estimations and often provide reliable uncertainty estimates that contribute to the robustness of the results. In this work, we propose a new metric to identify samples that are hard to classify. Our metric is defined as coincidence score for deep ensembles which measures the agreement of its individual models. The main hypothesis we rely on is that deep learning algorithms learn the low-loss samples better compared to large-loss samples. In order to compensate for this, we use controlled over-sampling on the identified ”hard” samples using proper data augmentation schemes to enable the models to learn those samples better. We validate the proposed metric using two public food datasets on different backbone architectures and show the improvements compared to the conventional deep neural network training using different performance metrics.     

Learning knowledge representation with meta knowledge distillation for single image super-resolution

Although the deep CNN-based super-resolution methods have achieved outstanding performance, their memory cost and computational complexity severely limit their practical employment. Knowledge distillation (KD), which can efficiently transfer knowledge from a cumbersome network (teacher) to a compact network (student), has demonstrated its advantages in some computer vision applications. The representation of knowledge is vital for knowledge transferring and student learning, which is generally defined in hand-crafted manners or uses the intermediate features directly. In this paper, we propose a model-agnostic meta knowledge distillation method under the teacher–student architecture for the single image super-resolution task. It provides a more flexible and accurate way to help teachers transmit knowledge in accordance with the abilities of students via knowledge representation networks (KRNets) with learnable parameters. Specifically, the texture-aware dynamic kernels are generated from local information to decompose the distillation problem into texture-wise supervision for further promoting the recovery quality of high-frequency details. In addition, the KRNets are optimized in a meta-learning manner to ensure the knowledge transferring and the student learning are beneficial to improving the reconstructed quality of the student. Experiments conducted on various single image super-resolution datasets demonstrate that our proposed method outperforms existing defined knowledge representation-related distillation methods and can help super-resolution algorithms achieve better reconstruction quality without introducing any extra inference complexity.     

Correlation of mercury concentrations in tree core and lichen samples in southeastern Louisiana

To determine if a correlation exists between mercury concentrations in lichens and wood from tree cores, tree core and lichen samples were taken from sites under mercury advisories and analyzed for mercury using a Bacharach 50B Mercury Analyzer System employing the Manual Cold Vapor Atomic Absorption EPA method 7471A. Wood from tree cores was dated using tree ring data. Wood samples from the cores were analyzed for various depths in the tree core (periods of time) to determine if variations occur throughout the cores. Preliminary data indicate that a strong correlation exists between mercury concentrations in tree core and lichen samples. Samples were analyzed for the 6-month period of summer 2003–winter 2004. A correlation coefficient of 0.882 was found between the tree core and lichen data sets. The derived correlations were used to estimate concentrations of mercury in tree rings for sites in lichen samples which were previously analyzed. These predicted values compared favorably to recently determined concentrations of Hg in tree cores. The strong correlation between lichen and tree core Hg concentrations suggests similar uptake mechanism for the two types of biota.