LEARNING PRECONDITIONS FOR PLANNING FROM PLAN TRACES AND HTN STRUCTURE

A great challenge in developing planning systems for practical applications is the difficulty of acquiring the domain information needed to guide such systems. This paper describes a way to learn some of that knowledge. More specifically, the following points are discussed. (1) We introduce a theoretical basis for formally defining algorithms that learn preconditions for Hierarchical Task Network (HTN) methods. (2) We describe Candidate Elimination Method Learner ( CaMeL ), a supervised, eager, and incremental learning process for preconditions of HTN methods. We state and prove theorems about CaMeL's soundness, completeness, and convergence properties. (3) We present empirical results about CaMeL's convergence under various conditions. Among other things, CaMeL converges the fastest on the preconditions of the HTN methods that are needed the most often. Thus CaMeL's output can be useful even before it has fully converged.     

Instance-Based Learning Algorithms

Storing and using specific instances improves the performance of several supervised learning algorithms. These include algorithms that learn decision trees, classification rules, and distributed networks. However, no investigation has analyzed algorithms that use only specific instances to solve incremental learning tasks. In this paper, we describe a framework and methodology, called instance-based learning, that generates classification predictions using only specific instances. Instance-based learning algorithms do not maintain a set of abstractions derived from specific instances. This approach extends the nearest neighbor algorithm, which has large storage requirements. We describe how storage requirements can be significantly reduced with, at most, minor sacrifices in learning rate and classification accuracy. While the storage-reducing algorithm performs well on several real-world databases, its performance degrades rapidly with the level of attribute noise in training instances. Therefore, we extended it with a significance test to distinguish noisy instances. This extended algorithm's performance degrades gracefully with increasing noise levels and compares favorably with a noise-tolerant decision tree algorithm.     

Combined application of Product Lifecycle and Software Configuration Management systems for ITER remote handling

The advantages of Product Lifecycle Management (PLM) systems are widely understood among the industry and hence a PLM system is already in use by International Thermonuclear Experimental Reactor (ITER) Organization (IO). However, with the increasing involvement of software in the development, the role of Software Configuration Management (SCM) systems have become equally important. The SCM systems can be useful to meet the higher demands on Safety Engineering (SE), Quality Assurance (QA), Validation and Verification (V&V) and Requirements Management (RM) of the developed software tools. In an experimental environment, such as ITER, the new remote handling requirements emerge frequently. This means the development of new tools or the modification of existing tools and the development of new remote handling procedures or the modification of existing remote handling procedures. PLM and SCM systems together can be of great advantage in the development and maintenance of such remote handling system. In this paper, we discuss how PLM and SCM systems can be integrated together and play their role during the development and maintenance of ITER remote handling system. We discuss the possibility to investigate such setup at DTP2 (Divertor Test Platform 2), which is the full scale mock-up facility to verify the ITER divertor remote handling and maintenance concepts.     

A Review and Empirical Evaluation of Feature Weighting Methods for a Class of Lazy Learning Algorithms

Many lazy learning algorithms are derivatives of the k-nearest neighbor (k-NN) classifier, which uses a distance function to generate predictions from stored instances. Several studies have shown that k-NN's performance is highly sensitive to the definition of its distance function. Many k-NN variants have been proposed to reduce this sensitivity by parameterizing the distance function with feature weights. However, these variants have not been categorized nor empirically compared. This paper reviews a class of weight-setting methods for lazy learning algorithms. We introduce a framework for distinguishing these methods and empirically compare them. We observed four trends from our experiments and conducted further studies to highlight them. Our results suggest that methods which use performance feedback to assign weight settings demonstrated three advantages over other methods: they require less pre-processing, perform better in the presence of interacting features, and generally require less training data to learn good settings. We also found that continuous weighting methods tend to outperform feature selection algorithms for tasks where some features are useful but less important than others.     

Secretion-and-capture cell-surface display for selection of target-binding proteins

This report describes a new cell-surface display system, the Secretion and Capture Technology (SECANT?) platform, which relies on in vivo biotinylation of the protein of interest followed by its capture on the avidinated surface of the parent cell. Cell sorting techniques are then used to isolate clones that display target-binding protein. A distinguishing feature of this method is its ability to display complex proteins, such as full-length immunoglobulin G (IgG) antibodies, on living cells. In this proof-of-concept study, Saccharomyces cerevisiae cells that displayed Herceptin IgG were isolated from a 10,000-fold excess of cells that displayed a lysozyme-binding antibody.     

Results of CMM standalone tests at DTP2

A full scale physical test facility, DTP2 (Divertor Test Platform 2) has been established in Finland for demonstrating and refining the Remote Handling (RH) equipment designs for ITER. The first prototype RH equipment at DTP2 is the Cassette Multifunctional Mover (CMM) equipped with the Second Cassette End Effector (SCEE) delivered to DTP2 in October 2008. The purpose is to prove that CMM/SCEE prototype can be used successfully for the 2nd cassette RH operations. At the end of F4E grant “DTP2 test facility operation and upgrade preparation”, the RH operations of the 2nd cassette were successfully demonstrated to the representatives of Fusion for Energy (F4E).During the grant the High Level Control (HLC) software developed at DTP2 was integrated with the CMM/SCEE hardware. The performance criteria of the CMM/SCEE equipment were defined based on the ‘EN ISO 9283 Manipulating industrial robots – Performance criteria and related test methods’ standard. Considerable improvement to the performance was achieved with the aid of compensation functions, which took into account the deflections and the compliance effects caused by the Divertor Cassette weighting 9000 kg. According to measurements the positioning error at the furthest point of the cassette was reduced from 80 mm to 5 mm.So far the 2nd cassette mock-up has been installed and removed already some tens of times. The reliability of the HLC software is sufficient to operate the CMM/SCEE all day without interruptions. Also the execution of the automatic RH processes with the overall RH system is reliable and repeatable in terms of accuracy and cycle time.These experiments provide a solid basis for investigating the RH system ability to perform fail-safe operations in various failure scenarios and to recover from them. The target of the continuing R&D is to find out a more complete set of functional and non-functional requirements for the RH system for Divertor Cassette maintenance to ensure an adequate level of requirements and procedures for ITER.     

Instance-based prediction of real-valued attributes

Instance-based representations have been applied to numerous classification tasks with some success. Most of these applications involved predicting a symbolic class based on observed attributes. This paper presents an instance-based method for predicting a numeric value based on observed attributes. We prove that, given enough instances, if the numeric values are generated by continuous functions with bounded slope, then the predicted values are accurate approximations of the actual values. We demonstrate the utility of this approach by comparing it with a standard approach for value prediction. The instance-based approach requires neither ad hoc parameters nor background knowledge.     

GOAL‐DRIVEN AUTONOMY FOR RESPONDING TO UNEXPECTED EVENTS IN STRATEGY SIMULATIONS

To operate autonomously in complex environments, an agent must monitor its environment and determine how to respond to new situations. To be considered intelligent, an agent should select actions in pursuit of its goals, and adapt accordingly when its goals need revision. However, most agents assume that their goals are given to them; they cannot recognize when their goals should change. Thus, they have difficulty coping with the complex environments of strategy simulations that are continuous, partially observable, dynamic, and open with respect to new objects. To increase intelligent agent autonomy, we are investigating a conceptual model for goal reasoning called Goal‐Driven Autonomy (GDA), which allows agents to generate and reason about their goals in response to environment changes. Our hypothesis is that GDA enables an agent to respond more effectively to unexpected events in complex environments. We instantiate the GDA model in ARTUE (A utonomous R esponse t o U nexpected E vents), a domain‐independent autonomous agent. We evaluate ARTUE on scenarios from two complex strategy simulations, and report on its comparative benefits and limitations. By employing goal reasoning, ARTUE outperforms an off‐line planner and a discrepancy‐based replanner on scenarios requiring reasoning about unobserved objects and facts and on scenarios presenting opportunities outside the scope of its current mission.