一个多元逐步回归的二叉树分类器的设计与实现

作者设计并实现了一个基于多变元逐步回归的二叉树分类器.在树结构和特征子集的选择中采用了穷举法,比有限制条件的选择更合理更优化.用 FORTRAN 语言实现的“遍历”二叉树,充分利用了 FORTRAN 处理可调数组的能力,并采取适当技巧,从而最大限度地利用了计算机内存.该通用分类器,可用来对任何具有统计数据的模式进行分类.在对白血球的分类中,取得了五类97%,六类92.2%的高识别率.

Neural Network Exploration Using Optimal Experiment Design

We consider the question "How should one act when the only goal is to learn as much as possible?" Building on the theoretical results of Fedorov [1972] and MacKay [1992], we apply techniques from Optimal Experiment Design (OED) to guide the query/action selection of a neural network learner. We demonstrate that these techniques allow the learner to minimize its generalization error by exploring its domain efficiently and completely. We conclude that, while not a panacea, OED-based query/action has much to offer, especially in domains where its high computational costs can be tolerated.

(Semi-)Predictive Discretization During Model Selection

In this paper, we present an approach to discretizing multivariate continuous data while learning the structure of a graphical model. We derive the joint scoring function from the principle of predictive accuracy, which inherently ensures the optimal trade-off between goodness of fit and model complexity (including the number of discretization levels). Using the so-called finest grid implied by the data, our scoring function depends only on the number of data points in the various discretization levels. Not only can it be computed efficiently, but it is also independent of the metric used in the continuous space. Our experiments with gene expression data show that discretization plays a crucial role regarding the resulting network structure.

Learning and Example Selection for Object and Pattern Detection

This thesis presents a learning based approach for detecting classes of objects and patterns with variable image appearance but highly predictable image boundaries. It consists of two parts. In part one, we introduce our object and pattern detection approach using a concrete human face detection example. The approach first builds a distribution-based model of the target pattern class in an appropriate feature space to describe the target's variable image appearance. It then learns from examples a similarity measure for matching new patterns against the distribution-based target model. The approach makes few assumptions about the target pattern class and should therefore be fairly general, as long as the target class has predictable image boundaries. Because our object and pattern detection approach is very much learning-based, how well a system eventually performs depends heavily on the quality of training examples it receives. The second part of this thesis looks at how one can select high quality examples for function approximation learning tasks. We propose an {em active learning} formulation for function approximation, and show for three specific approximation function classes, that the active example selection strategy learns its target with fewer data samples than random sampling. We then simplify the original active learning formulation, and show how it leads to a tractable example selection paradigm, suitable for use in many object and pattern detection problems.

Hierarchical Shape Description of Objects by Selection and Modification of Prototypes

An approach towards shape description, based on prototype modification and generalized cylinders, has been developed and applied to the object domains pottery and polyhedra: (1) A program describes and identifies pottery from vase outlines entered as lists of points. The descriptions have been modeled after descriptions by archeologists, with the result that identifications made by the program are remarkably consisten with those of the archeologists. It has been possible to quantify their shape descriptors, which are everyday terms in our language applied to many sorts of objects besides pottery, so that the resulting descriptions seem very natural. (2) New parsing strategies for polyhedra overcome some limitations of previous work. A special feature is that the processes of parsing and identification are carried out simultaneously.

Improving the Performance of Overlay Routing and P2P File Sharing using Selfish Neighbor Selection

A foundational issue underlying many overlay network applications ranging from routing to P2P file sharing is that of connectivity management, i.e., folding new arrivals into the existing mesh and re-wiring to cope with changing network conditions. Previous work has considered the problem from two perspectives: devising practical heuristics for specific applications designed to work well in real deployments, and providing abstractions for the underlying problem that are tractable to address via theoretical analyses, especially game-theoretic analysis. Our work unifies these two thrusts first by distilling insights gleaned from clean theoretical models, notably that under natural resource constraints, selfish players can select neighbors so as to efficiently reach near-equilibria that also provide high global performance. Using Egoist, a prototype overlay routing system we implemented on PlanetLab, we demonstrate that our neighbor selection primitives significantly outperform existing heuristics on a variety of performance metrics; that Egoist is competitive with an optimal, but unscalable full-mesh approach; and that it remains highly effective under significant churn. We also describe variants of Egoist's current design that would enable it to scale to overlays of much larger scale and allow it to cater effectively to applications, such as P2P file sharing in unstructured overlays, based on the use of primitives such as scoped-flooding rather than routing.

EGOIST: Overlay Routing Using Selfish Neighbor Selection

A foundational issue underlying many overlay network applications ranging from routing to P2P file sharing is that of connectivity management, i.e., folding new arrivals into an existing overlay, and re-wiring to cope with changing network conditions. Previous work has considered the problem from two perspectives: devising practical heuristics for specific applications designed to work well in real deployments, and providing abstractions for the underlying problem that are analytically tractable, especially via game-theoretic analysis. In this paper, we unify these two thrusts by using insights gleaned from novel, realistic theoretic models in the design of Egoist – a prototype overlay routing system that we implemented, deployed, and evaluated on PlanetLab. Using measurements on PlanetLab and trace-based simulations, we demonstrate that Egoist's neighbor selection primitives significantly outperform existing heuristics on a variety of performance metrics, including delay, available bandwidth, and node utilization. Moreover, we demonstrate that Egoist is competitive with an optimal, but unscalable full-mesh approach, remains highly effective under significant churn, is robust to cheating, and incurs minimal overhead. Finally, we discuss some of the potential benefits Egoist may offer to applications.

Implications of Selfish Neighbor Selection in Overlay Networks

In a typical overlay network for routing or content sharing, each node must select a fixed number of immediate overlay neighbors for routing traffic or content queries. A selfish node entering such a network would select neighbors so as to minimize the weighted sum of expected access costs to all its destinations. Previous work on selfish neighbor selection has built intuition with simple models where edges are undirected, access costs are modeled by hop-counts, and nodes have potentially unbounded degrees. However, in practice, important constraints not captured by these models lead to richer games with substantively and fundamentally different outcomes. Our work models neighbor selection as a game involving directed links, constraints on the number of allowed neighbors, and costs reflecting both network latency and node preference. We express a node's "best response" wiring strategy as a k-median problem on asymmetric distance, and use this formulation to obtain pure Nash equilibria. We experimentally examine the properties of such stable wirings on synthetic topologies, as well as on real topologies and maps constructed from PlanetLab and AS-level Internet measurements. Our results indicate that selfish nodes can reap substantial performance benefits when connecting to overlay networks composed of non-selfish nodes. On the other hand, in overlays that are dominated by selfish nodes, the resulting stable wirings are optimized to such great extent that even non-selfish newcomers can extract near-optimal performance through naive wiring strategies.

Object Detection at the Optimal Scale with Hidden State Shape Models

Hidden State Shape Models (HSSMs) [2], a variant of Hidden Markov Models (HMMs) [9], were proposed to detect shape classes of variable structure in cluttered images. In this paper, we formulate a probabilistic framework for HSSMs which provides two major improvements in comparison to the previous method [2]. First, while the method in [2] required the scale of the object to be passed as an input, the method proposed here estimates the scale of the object automatically. This is achieved by introducing a new term for the observation probability that is based on a object-clutter feature model. Second, a segmental HMM [6, 8] is applied to model the "duration probability" of each HMM state, which is learned from the shape statistics in a training set and helps obtain meaningful registration results. Using a segmental HMM provides a principled way to model dependencies between the scales of different parts of the object. In object localization experiments on a dataset of real hand images, the proposed method significantly outperforms the method of [2], reducing the incorrect localization rate from 40% to 15%. The improvement in accuracy becomes more significant if we consider that the method proposed here is scale-independent, whereas the method of [2] takes as input the scale of the object we want to localize.

Evaluating marker-guided treatment selection strategies

© 2014, The International Biometric Society.A potential venue to improve healthcare efficiency is to effectively tailor individualized treatment strategies by incorporating patient level predictor information such as environmental exposure, biological, and genetic marker measurements. Many useful statistical methods for deriving individualized treatment rules (ITR) have become available in recent years. Prior to adopting any ITR in clinical practice, it is crucial to evaluate its value in improving patient outcomes. Existing methods for quantifying such values mainly consider either a single marker or semi-parametric methods that are subject to bias under model misspecification. In this article, we consider a general setting with multiple markers and propose a two-step robust method to derive ITRs and evaluate their values. We also propose procedures for comparing different ITRs, which can be used to quantify the incremental value of new markers in improving treatment selection. While working models are used in step I to approximate optimal ITRs, we add a layer of calibration to guard against model misspecification and further assess the value of the ITR non-parametrically, which ensures the validity of the inference. To account for the sampling variability of the estimated rules and their corresponding values, we propose a resampling procedure to provide valid confidence intervals for the value functions as well as for the incremental value of new markers for treatment selection. Our proposals are examined through extensive simulation studies and illustrated with the data from a clinical trial that studies the effects of two drug combinations on HIV-1 infected patients.

Practical solvent system selection for counter-current separation of pharmaceutical compounds

Counter-current chromatography (CCC) is a technique that shows a lot of potential for large scale purification. Its usefulness in a "research and development" pharmaceutical environment has been investigated, and the conclusions are shown in this article. The use of CCC requires the development of an appropriate solvent system (a parameter of critical importance), a process which can be tedious. This article presents a novel strategy, combining a statistical approach and fast HPLC to generate a three-dimensional partition coefficient map and rapidly predict an optimal solvent system. This screen is performed in half a day and involves 9 experiments per solvent mixture. Test separations were performed using that screen to ensure the validity of the method.

A new procedure to optimize the selection of groups in a classification tree

Agglomerative cluster analyses encompass many techniques, which have been widely used in various fields of science. In biology, and specifically ecology, datasets are generally highly variable and may contain outliers, which increase the difficulty to identify the number of clusters. Here we present a new criterion to determine statistically the optimal level of partition in a classification tree. The criterion robustness is tested against perturbated data (outliers) using an observation or variable with values randomly generated. The technique, called Random Simulation Test (RST), is tested on (1) the well-known Iris dataset [Fisher, R.A., 1936. The use of multiple measurements in taxonomic problems. Ann. Eugenic. 7, 179–188], (2) simulated data with predetermined numbers of clusters following Milligan and Cooper [Milligan, G.W., Cooper, M.C., 1985. An examination of procedures for determining the number of clusters in a data set. Psychometrika 50, 159–179] and finally (3) is applied on real copepod communities data previously analyzed in Beaugrand et al. [Beaugrand, G., Ibanez, F., Lindley, J.A., Reid, P.C., 2002. Diversity of calanoid copepods in the North Atlantic and adjacent seas: species associations and biogeography. Mar. Ecol. Prog. Ser. 232, 179–195]. The technique is compared to several standard techniques. RST performed generally better than existing algorithms on simulated data and proved to be especially efficient with highly variable datasets.

Developing Approaches for Solving a Telecommunications Feature Subscription Problem

Call control features (e.g., call-divert, voice-mail) are primitive options to which users can subscribe off-line to personalise their service. The configuration of a feature subscription involves choosing and sequencing features from a catalogue and is subject to constraints that prevent undesirable feature interactions at run-time. When the subscription requested by a user is inconsistent, one problem is to find an optimal relaxation, which is a generalisation of the feedback vertex set problem on directed graphs, and thus it is an NP-hard task. We present several constraint programming formulations of the problem. We also present formulations using partial weighted maximum Boolean satisfiability and mixed integer linear programming. We study all these formulations by experimentally comparing them on a variety of randomly generated instances of the feature subscription problem.

Identifying optimal sites for natural recovery and restoration of impacted biogenic habitats in a special area of conservation using hydrodynamic and habitat suitability modelling

Selection of sites for successful restoration of impacted shellfish populations depends on understanding the dispersion capability and habitat requirements of the species involved. In Strangford Lough, Northern Ireland, the horse mussel (Modiolus modiolus) biogenic reefs cover only a fraction of their historical range with the remaining reefs badly damaged and requiring restoration. Previous experimental trials suggest that translocation of horse mussels accelerates reef recovery and has therefore been proposed as a suitable restoration technique. We used a series of coupled hydrodynamic and particle dispersal models to assess larval dispersion from remnant and translocated populations to identify suitable areas for adult live M. modiolus translocation in Strangford Lough, Northern Ireland. A maximum entropy model (MAXENT) was used to identify if dispersing larvae could reach habitat suitable for adult M. modiolus. From these we predicted if translocated mussels will reseed themselves or be able to act as larval sources for nearby reefs. The dispersal models showed that the remnant M. modiolus populations are largely self-recruiting with little connectivity between them. The majority of larvae settled near the sources and movement was largely dependent on the tides and not influenced by wind or waves. Higher reef elevation resulted in larvae being able to disperse further away from the release point. However, larval numbers away from the source population are likely to be too low for successful recruitment. There was also little connectivity between the Irish Sea and Strangford Lough as any larvae entering the Lough remained predominantly in the Strangford Narrows. The areas covered by these self-seeding populations are suitable for M. modiolus translocation according to the MAXENT model. As a result of this work and in conjunction with other field work we propose a combination of total protection of all remaining larval sources and small scale translocations onto suitable substrata in each of the identified self-recruiting areas.

Impact of Metric Selection on Wireless DeAuthentication DoS Attack Performance

DeAuthentication Denial of Service attacks in Public Access WiFi operate by exploiting the lack of authentication of management frames in the 802.11 protocol. Detection of these attacks rely almost exclusively on the selection of appropriate thresholds. In this work the authors demonstrate that there are additional, previously unconsidered, metrics which also influence DoS detection performance. A method of systematically tuning these metrics to optimal values is proposed which ensures that parameter choices are repeatable and verifiable.