자연주의적 탐구조사(Naturalistic Inquiry)

THE DESIGN AND ANALYSIS OF EXPERIMENTS. By Oscar Kempthorne. New York, John Wiley and Sons, Inc., 1952. 631 pp. $8.50. This book by a teacher of statistics (as well as a consultant for \"experimenters\") is a comprehensive study of the philosophical background for the statistical design of experiment. It is necessary to have some facility with algebraic notation and manipulation to be able to use the volume intelligently. The problems are presented from the theoretical point of view, without such practical examples as would be helpful for those not acquainted with mathematics. The mathematical justification for the techniques is given. As a somewhat advanced treatment of the design and analysis of experiments, this volume will be interesting and helpful for many who approach statistics theoretically as well as practically. With emphasis on the \"why,\" and with description given broadly, the author relates the subject matter to the general theory of statistics and to the general problem of experimental inference. MARGARET J. ROBERTSON

The Design and Analysis of Experiments

Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

Data Mining Practical Machine Learning Tools and Techniques

In many engineering optimization problems, the number of function evaluations is severely limited by time or cost. These problems pose a special challenge to the field of global optimization, since existing methods often require more function evaluations than can be comfortably afforded. One way to address this challenge is to fit response surfaces to data collected by evaluating the objective and constraint functions at a few points. These surfaces can then be used for visualization, tradeoff analysis, and optimization. In this paper, we introduce the reader to a response surface methodology that is especially good at modeling the nonlinear, multimodal functions that often occur in engineering. We then show how these approximating functions can be used to construct an efficient global optimization algorithm with a credible stopping rule. The key to using response surfaces for global optimization lies in balancing the need to exploit the approximating surface (by sampling where it is minimized) with the need to improve the approximation (by sampling where prediction error may be high). Striking this balance requires solving certain auxiliary problems which have previously been considered intractable, but we show how these computational obstacles can be overcome.

/pdf/efficient-global-optimization-of-expensive-black-box-4m8fv6oubz.pdf

Efficient Global Optimization of Expensive Black-Box Functions

E-mail and World Wide Web surveys have been the subject of much hyperbole about their capabilities as well as some criticism of their limitations. In this report, the authors examine what is known ...

/pdf/advantages-and-disadvantages-of-internet-research-surveys-irllqkux8r.pdf

Advantages and Disadvantages of Internet Research Surveys: Evidence from the Literature:

BACKGROUND. The use of androgen deprivation therapy (ADT) in the treatment of men with prostate cancer has risen sharply. Although cardiovascular disease is the most common reason for death among men with prostate cancer who do not die of the disease itself, data regarding the effect of ADT on cardiovascular morbidity and mortality in men with prostate cancer are limited. In the current study, the authors attempted to measure the risk for subsequent cardiovascular morbidity in men with prostate cancer who received ADT. METHODS. A cohort of newly diagnosed men in a population-based registry who were diagnosed between 1992 and 1996 were identified retrospectively. A total of 22,816 subjects were identified after exclusion criteria were applied. Using a multivariate model, the authors calculated the risk of subsequent cardiovascular morbidity in men with prostate cancer who were treated with ADT, as defined using Medicare claims. RESULTS. Newly diagnosed prostate cancer patients who received ADT for at least 1 year were found to have a 20% higher risk of serious cardiovascular morbidity compared with similar men who did not receive ADT. Subjects began incurring this higher risk within 12 months of treatment. However, Hispanic men were found to have a lowered risk for cardiovascular morbidity. CONCLUSIONS. ADT is associated with significantly increased cardiovascular morbidity in men with prostate cancer and may lower overall survival in men with low-risk disease. These data have particular relevance to decisions regarding the use of ADT in men with prostate cancer in settings in which the benefit has not been clearly established. For men with metastatic disease, focused efforts to reduce cardiac risk factors through diet, exercise, or the use of lipid-lowering agents may mitigate some of the risks of ADT. Cancer 2007;110:1493–500. � 2007 American Cancer Society.

/pdf/androgen-deprivation-therapy-increases-cardiovascular-265bxq76sb.pdf

Androgen deprivation therapy increases cardiovascular morbidity in men with prostate cancer.

/pdf/global-versus-local-search-in-constrained-optimization-of-421d5nx9qr.pdf

Global versus local search in constrained optimization of computer models

A complex mathematical model that produces output values from input values is now commonly called a computer model. This thesis considers the problem of finding the global optimum of the response with few function evaluations. A small number of function evaluations is desirable since the computer model is often expensive (time consuming) to evaluate.
The function to be optimized is modeled as a stochastic process from initial function evaluations. Points are sampled sequentially according to a criterion that combines promising prediction values with prediction uncertainty. Some graphical tools are given that allow early assessment about whether the modeling strategy will work well. The approach is generalized by introducing a parameter that controls how global versus local the search strategy is. Strategies to conduct the optimization in stages and for optimization subject to constraints on additional response variables are presented.
Special consideration is given to the stopping criterion of the global optimization algorithm. The problem of achieving a tolerance on the global minimum can be represented by determining whether the first order statistic of N dependent variables is greater than a certain value. An algorithm is developed that quickly determines bounds on the probability of this event.
A strategy to explore high-dimensional data informally through effect plots is presented. The interpretation of the plots is guided by pointwise standard errors of the effects which are developed. When used in the context of global optimization, the graphical analysis sheds light on the number and location of local optima.

Matthias Schonlau

Papers

Efficient Global Optimization of Expensive Black-Box Functions

Advantages and Disadvantages of Internet Research Surveys: Evidence from the Literature:

Androgen deprivation therapy increases cardiovascular morbidity in men with prostate cancer.

Global versus local search in constrained optimization of computer models

Computer experiments and global optimization