Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

According to W. Edwards Deming, American companies require nothing less than a transformation of management style and of governmental relations with industry. In Out of the Crisis, originally published in 1982, Deming offers a theory of management based on his famous 14 Points for Management. Management's failure to plan for the future, he claims, brings about loss of market, which brings about loss of jobs. Management must be judged not only by the quarterly dividend, but by innovative plans to stay in business, protect investment, ensure future dividends, and provide more jobs through improved product and service. In simple, direct language, he explains the principles of management transformation and how to apply them.

Out of the Crisis

The concept of an agent has become important in both Artificial Intelligence (AI) and mainstream computer science. Our aim in this paper is to point the reader at what we perceive to be the most important theoretical and practical issues associated with the design and construction of intelligent agents. For convenience, we divide these issues into three areas (though as the reader will see, the divisions are at times somewhat arbitrary). Agent theory is concerned with the question of what an agent is, and the use of mathematical formalisms for representing and reasoning about the properties of agents. Agent architectures can be thought of as software engineering models of agents;researchers in this area are primarily concerned with the problem of designing software or hardware systems that will satisfy the properties specified by agent theorists. Finally, agent languages are software systems for programming and experimenting with agents; these languages may embody principles proposed by theorists. The paper is not intended to serve as a tutorial introduction to all the issues mentioned; we hope instead simply to identify the most important issues, and point to work that elaborates on them. The article includes a short review of current and potential applications of agent technology.

/pdf/intelligent-agents-theory-and-practice-5dchj7p8pc.pdf

Intelligent Agents: Theory and Practice

物件導向軟體之架構(Object-Oriented Software Construction)探討

From the Publisher: 
The accessible presentation of this book gives both a general view of the entire computer vision enterprise and also offers sufficient detail to be able to build useful applications. Users learn techniques that have proven to be useful by first-hand experience and a wide range of mathematical methods. A CD-ROM with every copy of the text contains source code for programming practice, color images, and illustrative movies. Comprehensive and up-to-date, this book includes essential topics that either reflect practical significance or are of theoretical importance. Topics are discussed in substantial and increasing depth. Application surveys describe numerous important application areas such as image based rendering and digital libraries. Many important algorithms broken down and illustrated in pseudo code. Appropriate for use by engineers as a comprehensive reference to the computer vision enterprise.

Computer vision : a modern approach = 计算机视觉 : 一种现代的方法

We present several general, broadly applicable mechanisms that enable computations to execute with reduced resources, typically at the cost of some loss in the accuracy of the result they produce.We identify several general computational patterns that interact well with these resource reduction mechanisms, present a concrete manifestation of these patterns in the form of simple model programs, perform simulationbased explorations of the quantitative consequences of applying these mechanisms to our model programs, and relate the model computations (and their interaction with the resource reduction mechanisms) to more complex benchmark applications drawn from a variety of fields.

/pdf/patterns-and-statistical-analysis-for-understanding-reduced-3wc86jnhao.pdf

Patterns and statistical analysis for understanding reduced resource computing

In this position paper, we describe our vision of the future of machine programming through a categorical examination of three pillars of research. Those pillars are: (i) intention, (ii) invention, and(iii) adaptation. Intention emphasizes advancements in the human-to-computer and computer-to-machine-learning interfaces. Invention emphasizes the creation or refinement of algorithms or core hardware and software building blocks through machine learning (ML). Adaptation emphasizes advances in the use of ML-based constructs to autonomously evolve software.

The Three Pillars of Machine Programming

An acceptability envelope is a region of imperfect but acceptable software systems surrounding a given perfect system. Explicitly targeting the acceptability envelope during development (rather than attempting to minimize the number of errors, as is the current practice) has several potential benefits. Specifically, leaving acceptable errors in the system eliminates the risks and costs associated with attempting to repair the errors; investing fewer resources in less critical regions of the program and more resources in more critical regions may increase acceptability and reduce the overall investment of development resources.To realize these benefits, the acceptability envelope must be both sizable and accessible. We present several case studies that explore the acceptability envelopes of the Pine email client and the Sure-Player MPEG decoder. These studies show that both Pine and Sure-Player can tolerate the addition of many off-by-one errors without producing unacceptable behavior. This result suggests that current systems may be overengineered in the sense that they can tolerate many more errors than they currently contain.Our SurePlayer case study also shows that SurePlayer has unforgiving regions of code that must be close to perfect for the system to function at all. To effectively exploit the acceptability envelope, developers must be able to distinguish forgiving and unforgiving regions so that they can appropriately prioritize their development effort. In SurePlayer, the unforgiving regions occur in code that uses metadata to parse the input stream; the forgiving regions tend to access the data within each image. This result suggests that developers may be able to use relatively simple indicators to effectively prioritize their development effort.

/pdf/exploring-the-acceptability-envelope-32dl56xnai.pdf

Exploring the acceptability envelope

We present Bolt, a novel system for escaping from infinite and long-running loops. Directed by a user, Bolt can attach to a running process and determine if the program is executing an infinite loop. If so, Bolt can deploy multiple strategies to escape the loop, restore the responsiveness of the program, and enable the program to deliver useful output.Bolt operates on stripped x86 and x64 binaries, dynamically attaches and detaches to and from the program as needed, and dynamically detects loops and creates program state checkpoints to enable exploration of different escape strategies. Bolt can detect and escape from loops in off-the-shelf software, without available source code, and with no overhead in standard production use.

/pdf/bolt-on-demand-infinite-loop-escape-in-unmodified-binaries-56wqhyqydh.pdf

Bolt: on-demand infinite loop escape in unmodified binaries

Modern software pervasively uses structurally complex data such as linked data structures. The standard approach to generating test suites for such software, manual generation of the inputs in the suite, is tedious and error-prone. This dissertation proposes a new approach for generating suites with structurally complex test inputs; presents a technique that automates this approach; describes the Korat tool that implements this technique for Java; and evaluates the effectiveness of Korat in testing a set of data-structure implementations. 
Our approach allows the developer to describe the properties of valid test inputs using a familiar implementation language such as Java. Specifically, the user provides an imperative predicate—a piece of code that returns a truth value—that returns true if the input satisfies the required property and false otherwise. Korat implements our technique for solving imperative predicates: given a predicate and a bound on the size of the predicate's inputs, Korat automatically generates the bounded-exhaustive test suite that consists of all inputs, within the given bound, that satisfy the property identified by the predicate. To generate these inputs, Korat systematically searches the bounded input space by executing the predicate on the candidate inputs. Korat does this efficiently by pruning the search based on the predicate's executions and by generating only non-isomorphic inputs. 
Bounded-exhaustive testing is a methodology for testing the code on all inputs within the given small bound. Our experiments on a set of ten linked and array-based data structures show that Korat can efficiently generate bounded-exhaustive test suites from imperative predicates even for very large input spaces and that these test suites can achieve high statement, branch, and mutation coverage. Moreover, the use of our technique for generating structurally complex test inputs has enabled testers in industry to detect faults in real, production-quality applications. (Copies available exclusively from MIT Libraries, Rm. 14-0551, Cambridge, MA 02139-4307. Ph. 617-253-5668; Fax 617-253-1690.)

Martin Rinard

Papers

Patterns and statistical analysis for understanding reduced resource computing

The Three Pillars of Machine Programming

Exploring the acceptability envelope

Bolt: on-demand infinite loop escape in unmodified binaries

Automatic testing of software with structurally complex inputs