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 - Concepts and Techniques.

Planning algorithms are impacting technical disciplines and industries around the world, including robotics, computer-aided design, manufacturing, computer graphics, aerospace applications, drug design, and protein folding. This coherent and comprehensive book unifies material from several sources, including robotics, control theory, artificial intelligence, and algorithms. The treatment is centered on robot motion planning but integrates material on planning in discrete spaces. A major part of the book is devoted to planning under uncertainty, including decision theory, Markov decision processes, and information spaces, which are the “configuration spaces” of all sensor-based planning problems. The last part of the book delves into planning under differential constraints that arise when automating the motions of virtually any mechanical system. Developed from courses taught by the author, the book is intended for students, engineers, and researchers in robotics, artificial intelligence, and control theory as well as computer graphics, algorithms, and computational biology.

Planning Algorithms: Introductory Material

A new motion planning method for robots in static workspaces is presented. This method proceeds in two phases: a learning phase and a query phase. In the learning phase, a probabilistic roadmap is constructed and stored as a graph whose nodes correspond to collision-free configurations and whose edges correspond to feasible paths between these configurations. These paths are computed using a simple and fast local planner. In the query phase, any given start and goal configurations of the robot are connected to two nodes of the roadmap; the roadmap is then searched for a path joining these two nodes. The method is general and easy to implement. It can be applied to virtually any type of holonomic robot. It requires selecting certain parameters (e.g., the duration of the learning phase) whose values depend on the scene, that is the robot and its workspace. But these values turn out to be relatively easy to choose, Increased efficiency can also be achieved by tailoring some components of the method (e.g., the local planner) to the considered robots. In this paper the method is applied to planar articulated robots with many degrees of freedom. Experimental results show that path planning can be done in a fraction of a second on a contemporary workstation (/spl ap/150 MIPS), after learning for relatively short periods of time (a few dozen seconds).

/pdf/probabilistic-roadmaps-for-path-planning-in-high-dimensional-45s4l7x8cl.pdf

Probabilistic roadmaps for path planning in high-dimensional configuration spaces

During the last decade, sampling-based path planning algorithms, such as probabilistic roadmaps (PRM) and rapidly exploring random trees (RRT), have been shown to work well in practice and possess theoretical guarantees such as probabilistic completeness. However, little effort has been devoted to the formal analysis of the quality of the solution returned by such algorithms, e.g. as a function of the number of samples. The purpose of this paper is to fill this gap, by rigorously analyzing the asymptotic behavior of the cost of the solution returned by stochastic sampling-based algorithms as the number of samples increases. A number of negative results are provided, characterizing existing algorithms, e.g. showing that, under mild technical conditions, the cost of the solution returned by broadly used sampling-based algorithms converges almost surely to a non-optimal value. The main contribution of the paper is the introduction of new algorithms, namely, PRM* and RRT*, which are provably asymptotically optimal, i.e. such that the cost of the returned solution converges almost surely to the optimum. Moreover, it is shown that the computational complexity of the new algorithms is within a constant factor of that of their probabilistically complete (but not asymptotically optimal) counterparts. The analysis in this paper hinges on novel connections between stochastic sampling-based path planning algorithms and the theory of random geometric graphs.

Sampling-based algorithms for optimal motion planning

Several randomized path planners have been proposed during the last few years. Their attractiveness stems from their applicability to virtually any type of robots, and their empirically observed success. In this paper we attempt to present a unifying view of these planners and to theoretically explain their success. First, we introduce a general planning scheme that consists of randomly sampling the robot’s configuration space. We then describe two previously developed planners as instances of planners based on this scheme, but applying very different sampling strategies. These planners are probabilistically complete: if a path exists, they will find one with high probability, if we let them run long enough. Next, for one of the planners, we analyze the relation between the probability of failure and the running time. Under assumptions characterizing the “goodness” of the robot’s free space, we show that the running time only grows as the absolute value of the logarithm of the probability of failure that we are willing to tolerate. We also show that it increases at a reasonable rate as the space goodness degrades. In the last section we suggest directions for future research.

/pdf/a-random-sampling-scheme-for-path-planning-2tkhl0aaj4.pdf

A random sampling scheme for path planning

Maintainability is an important issue in design where the accessibility of certain parts is determined for routine maintenance. In the past its study has been largely manual and labor intensive. Either by using physical mockup or computer animation with CAD models of a design, the task relies on a human to provide an access path for the part. In this paper, the authors present an automated approach to replace this manual process. By applying results from and developing extensions to research in motion planning and other fields, the authors demonstrate that an automated maintainability study system is feasible. The authors describe general extensions needed to adapt robotic motion planning techniques in maintainability studies. The authors show results from applying such a system to two classes of industrial application problems.

http://www3.nccu.edu.tw/~li/Publication/pdf/icra95b.pdf

Assembly maintainability study with motion planning

Traditional approaches to the motion-planning problem can be classified into single-query and multiple-query problems with the tradeoffs on run-time computation cost and adaptability to environment changes. In this paper, we propose a novel approach to the problem that can learn incrementally on every planning query and effectively manage the learned roadmap as the process goes on. This planner is based on previous work on probabilistic roadmaps and uses a data structure, called reconfigurable random forest (RRF), which extends the rapidly-exploring random tree structure proposed in the literature. The planner can account for environmental changes while keeping the size of the roadmap small. The planner removes invalid nodes in the roadmap as the obstacle configurations change. It also uses a tree-pruning algorithm to trim RRF into a more concise representation. Our experiments show that the planner is flexible and efficient.

http://www3.nccu.edu.tw/~li/Publication/pdf/icra2002.pdf

An incremental learning approach to motion planning with roadmap management

The work aims to improve the assessment of creative problem-solving in science education by employing language technologies and computational-statistical machine learning methods to grade students' natural language responses automatically. To evaluate constructs like creative problem-solving with validity, open-ended questions that elicit students' constructed responses are beneficial. But the high cost required in manually grading constructed responses could become an obstacle in applying open-ended questions. In this study, automated grading schemes have been developed and evaluated in the context of secondary Earth science education. Empirical evaluations revealed that the automated grading schemes may reliably identify domain concepts embedded in students' natural language responses with satisfactory inter-coder agreement against human coding in two sub-tasks of the test (Cohen's Kappa=.65-.72). And when a single holistic score was computed for each student, machine-generated scores achieved high inter-rater reliability against human grading (Pearson's r=.92). The reliable performance in automatic concept identification and numeric grading demonstrates the potential of using automated grading to support the use of open-ended questions in science assessments and enable new technologies for science learning.

http://www3.nccu.edu.tw/~li/Publication/pdf/cae2008.pdf

Assessing creative problem-solving with automated text grading

We developed a system for automatic generation of caricatures by analyzing unique facial features of the subject. The proposed face model contains 119 parameters derived from the MPEG-4 standard. These nodes are categorized into 8 groups with a pre-defined hierarchy to guide the placement of face components after shape exaggeration. Given an input image, the system determines which and how the face component should be altered. Using an artist’s finished work as the source image, the system is capable of producing caricatures of a similar style effectively and efficiently.

http://www3.nccu.edu.tw/~li/Publication/pdf/accv2004.pdf

Tsai-Yen Li

Papers

A random sampling scheme for path planning

Assembly maintainability study with motion planning

An incremental learning approach to motion planning with roadmap management

Assessing creative problem-solving with automated text grading

Automatic Caricature Generation by Analyzing Facial Features