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

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

This paper studies the control of a class of discrete event processes, i.e. processes that are discrete, asynchronous and possibly nondeter-ministic. The controlled process is described as the generator of a formal language, while the controller, or supervisor, is constructed from the grammar of a specified target language that incorporates the desired closed-loop system behavior. The existence problem for a supervisor is reduced to finding the largest controllable language contained in a given legal language. Two examples are provided.

/pdf/supervisory-control-of-a-class-of-discrete-event-processes-45fsmhdhbo.pdf

Supervisory control of a class of discrete event processes

Since the subject of traffic dynamics has captured the interest of physicists, many surprising effects have been revealed and explained. Some of the questions now understood are the following: Why are vehicles sometimes stopped by ``phantom traffic jams'' even though drivers all like to drive fast? What are the mechanisms behind stop-and-go traffic? Why are there several different kinds of congestion, and how are they related? Why do most traffic jams occur considerably before the road capacity is reached? Can a temporary reduction in the volume of traffic cause a lasting traffic jam? Under which conditions can speed limits speed up traffic? Why do pedestrians moving in opposite directions normally organize into lanes, while similar systems ``freeze by heating''? All of these questions have been answered by applying and extending methods from statistical physics and nonlinear dynamics to self-driven many-particle systems. This article considers the empirical data and then reviews the main approaches to modeling pedestrian and vehicle traffic. These include microscopic (particle-based), mesoscopic (gas-kinetic), and macroscopic (fluid-dynamic) models. Attention is also paid to the formulation of a micro-macro link, to aspects of universality, and to other unifying concepts, such as a general modeling framework for self-driven many-particle systems, including spin systems. While the primary focus is upon vehicle and pedestrian traffic, applications to biological or socio-economic systems such as bacterial colonies, flocks of birds, panics, and stock market dynamics are touched upon as well.

Traffic and related self-driven many-particle systems

https://www.ufz.de/export/data/2/100065_GrimmODD.pdf

A standard protocol for describing individual-based and agent-based models

A model can be decomposed into components, then mapped onto a network of simulators for execution. With different levels of decompositions, we are interested in finding the optimal model decomposition that can be implemented on a particular structure of distributed simulators. This is done by comparing performance measures such as the minimum response time or the maximum throughout per unit of hardware complexity. We claim that the need for a model capable of performance evaluation is an important concern in designing distributed simulators. Thus a methodology for performance evaluation and simulation modelling of hierarchically specified distributed simulators is required. From the empirical results, we found that simulating a given distributed simulator model is suitable to get optimal performances, that is, each model has an optimal decomposition level in terms of performance. The influential performance measures are the average of simulation run times and the product of the average run time and the hardware complexity of the given simulator model.

/pdf/performance-evaluation-of-hierarchical-distributed-xeesx6gl83.pdf

Performance evaluation of hierarchical distributed simulators

DEVS/NS-2 is a new integrated network simulation tool that provides efficient high and low level network modeling capabilities. DEVS (Discrete Event System Specification) serves as a base framework to support well-defined mathematical formalism specification for structure and behavior of dynamic systems. The NS-2 is a discrete event network simulator, whose primary use is intended to build and run various detailed network models and protocols. The interoperable simulation of the two powerful modeling and simulation systems, DEVS and NS-2, allows to attain significant benefits in terms of reduction of the cost, increased high and low level networks modeling power, and enhanced reusability.

DEVS/NS-2 environment: integrated tool for efficient networks modeling and simulation

Knowledge-based simulation design methodology. A flexible test architecture application

This paper presents a new extension of the DEVS formalism that allows multiple occurrences of a given instance of a DEVS component. This paper is a follow-up to a previous short paper in which the issue of supporting a new construction called a shared component was raised, in the case of a DEVS model. In this paper, we first demonstrate, formally, that the multi-occurrence extended definition, that includes the case of shared components, is valid because any model that is built using this extended definition accepts an equivalent model built using standard DEVS. Then we recall the benefits of sharing components for modeling, and further extend this analysis to the simulation area, by investigating how shared components can help to design better simulation engines. Finally, we describe an existing implementation of a simulation software that fully supports this shared component feature, both at the modeling and simulation levels.

/pdf/extending-devs-to-support-multiple-occurrence-in-component-41anrw8wkz.pdf

Extending DEVS to support multiple occurrence in component-based simulation

In Discrete Event System Specification (DEVS), the dynamics of a network is constituted only by the dynamics of its basic components. The state of each component is fully encapsulated. Control in the network is fully decentralized to each component. At dynamic structure level, DEVS should permit the same level of decentralization. However, it is hard to ensure structure consistency while letting all components achieve structure changes. Besides, this solution can be complex to implement. To avoid these difficulties, usual dynamic structure approaches ensure structure consistency allowing structure changes to be done only by the network having newly added dynamics change capabilities. This is a safe and simple way to achieve dynamic structure. However, it should be possible to simply allow components of a network to modify the structure of their network, other components and/or their own structure — without having to modify the usual definition a DEVS network. In this manuscript, it is shown that a simple fully decentralized approach is possible while ensuring full modularity and structure consistency.

/pdf/specification-of-dynamic-structure-discrete-event-systems-3nhenlephv.pdf

Bernard P. Zeigler

Papers

Performance evaluation of hierarchical distributed simulators

DEVS/NS-2 environment: integrated tool for efficient networks modeling and simulation

Knowledge-based simulation design methodology. A flexible test architecture application

Extending DEVS to support multiple occurrence in component-based simulation

Specification of dynamic structure discrete event systems using single point encapsulated control functions