Diverse reasoning in automated model formulation

Diverse reasoning supports a dynamic integration of various reasoning methods in a computerized system. This paper describes a control blackboard approach to simulate the control features observed in the expert's model formulation protocols. The diverse reasoning concept is incorporated so that the model formulation process is dynamically in a plan-directed, action-directed, or data-directed fashion. The diverse reasoning concept facilitates the control features simulation. By analyzing the diverse reasoning behavior in the proposed system, this paper contributes to a better understanding of and support to the modeling process for the design of intelligent decision support systems. The usefulness of the prototype system is also evaluated using an empirical experiment.     

Boosting for transfer learning from multiple data sources

Transfer learning aims at adapting a classifier trained on one domain with adequate labeled samples to a new domain where samples are from a different distribution and have no class labels. In this paper, we explore the transfer learning problems with multiple data sources and present a novel boosting algorithm, SharedBoost. This novel algorithm is capable of applying for very high dimensional data such as in text mining where the feature dimension is beyond several ten thousands. The experimental results illustrate that the SharedBoost algorithm significantly outperforms the traditional methods which transfer knowledge with supervised learning techniques. Besides, SharedBoost also provides much better classification accuracy and more stable performance than some other typical transfer learning methods such as the structural correspondence learning (SCL) and the structural learning in the multiple sources transfer learning problems.     

On universal transfer learning

In transfer learning the aim is to solve new learning tasks using fewer examples by using information gained from solving related tasks. Existing transfer learning methods have been used successfully in practice and PAC analysis of these methods have been developed. But the key notion of relatedness between tasks has not yet been defined clearly, which makes it difficult to understand, let alone answer, questions that naturally arise in the context of transfer, such as, how much information to transfer, whether to transfer information, and how to transfer information across tasks. In this paper, we look at transfer learning from the perspective of Algorithmic Information Theory/Kolmogorov complexity theory, and formally solve these problems in the same sense Solomonoff Induction solves the problem of inductive inference. We define universal measures of relatedness between tasks, and use these measures to develop universally optimal Bayesian transfer learning methods. We also derive results in AIT that are interesting by themselves. To address a concern that arises from the theory, we also briefly look at the notion of Kolmogorov complexity of probability measures. Finally, we present a simple practical approximation to the theory to do transfer learning and show that even these are quite effective, allowing us to transfer across tasks that are superficially unrelated. The latter is an experimental feat which has not been achieved before, and thus shows the theory is also useful in constructing practical transfer algorithms.     

Variant design for mechanical artifacts: A state-of-the-art survey

Variant design refers, to the technique of adapting existing design specifications to satisfy new design goals and constraints. Specific support of variant design techniques in current computer aided design systems would help to realize a rapid response manufacturing environment. A survey of approaches supporting variant design is presented. Capabilities used in current commercial computer aided design systems are discussed along with approaches used in recent research efforts. Information standards applicable to variant design are also identified. Barriers to variant design in current systems are identified and ideas are presented for augmentation of current systems to support variant design.     

Experience transfer for process improvement

The oil well drilling process is the selected representative of a challenging industrial process. The drilling process is becoming more complex as oil fields mature and technology evolves. At the same time, the amount of information is increasing in volume and frequency. Although technology is advancing, failures occur at almost the same rate as before, leading to loss of valuable time. Whenever the process is failing, or running smoothly, valuable experience is gained. To take advantage of established and continually growing new experience a formalized methodology, knowledge intensive case-based reasoning, was applied for capturing of drilling process experience and for reusing it. Experience was collected from different information sources. Structured cases were used to describe failure episodes; its circumstances and how the failure was repaired. A general domain knowledge model supports the case-based reasoning process. It was demonstrated how the system was able to recommend how to solve problems when they arise, while at the same time bridging the gap between new and experienced personnel. Method performance was tested on 62 selected field cases. The system also identified the failure causes of problems and could thereby suggest more effective repair actions.     

Flexible latent variable models for multi-task learning

Given multiple prediction problems such as regression or classification, we are interested in a joint inference framework that can effectively share information between tasks to improve the prediction accuracy, especially when the number of training examples per problem is small. In this paper we propose a probabilistic framework which can support a set of latent variable models for different multi-task learning scenarios. We show that the framework is a generalization of standard learning methods for single prediction problems and it can effectively model the shared structure among different prediction tasks. Furthermore, we present efficient algorithms for the empirical Bayes method as well as point estimation. Our experiments on both simulated datasets and real world classification datasets show the effectiveness of the proposed models in two evaluation settings: a standard multi-task learning setting and a transfer learning setting.     

Using background knowledge in case-based legal reasoning: A computational model and an intelligent learning environment

Researchers in the field of AI and Law have developed a number of computational models of the arguments that skilled attorneys make based on past cases. However, these models have not accounted for the ways that attorneys use middle-level normative background knowledge (1) to organize multi-case arguments, (2) to reason about the significance of differences between cases, and (3) to assess the relevance of precedent cases to a given problem situation. We present a novel model, that accounts for these argumentation phenomena. An evaluation study showed that arguments about the significance of distinctions based on this model help predict the outcome of cases in the area of trade secrets law, confirming the quality of these arguments. The model forms the basis of an intelligent learning environment called CATO, which was designed to help beginning law students acquire basic argumentation skills. CATO uses the model for a number of purposes, including the dynamic generation of argumentation examples. In a second evaluation study, carried out in the context of an actual legal writing course, we compared instruction with CATO against the best traditional legal writing instruction. The results indicate that CATO's example-based instructional approach is effective in teaching basic argumentation skills. However, a more “integrated” approach appears to be needed if students are to achieve better transfer of these skills to more complex contexts. CATO's argumentation model and instructional environment are a contribution to the research fields of AI and Law, Case-Based Reasoning, and AI and Education.     

A reduction-graph model of precedent in legal analysis

Legal analysis is a task underlying many forms of legal problem solving. In the Anglo-American legal system, legal analysis is based in part on legal precedents, previously decided cases. This paper describes a reduction-graph model of legal precedents that accounts for a key characteristic of legal precedents: a precedent's relevance to subsequent cases is determined by the theory under which the precedent is decided. This paper identifies the implementation requirements for legal analysis using the reduction-graph model of legal precedents and describes GREBE, a program that satisfies these requirements.     

An application of case-based reasoning with machine learning for forensic autopsy

Case-based reasoning (CBR) is one of the matured paradigms of artificial intelligence for problem solving. CBR has been applied in many areas in the commercial sector to assist daily operations. However, CBR is relatively new in the field of forensic science. Even though forensic personnel have consciously used past experiences in solving new cases, the idea of applying machine intelligence to support decision-making in forensics is still in its infancy and poses a great challenge. This paper highlights the limitation of the methods used in forensics compared with a CBR method in the analysis of forensic evidences. The design and development of an Intelligent Forensic Autopsy Report System (I-AuReSys) basing on a CBR method along with the experimental results are presented. Our system is able to extract features by using an information extraction (IE) technique from the existing autopsy reports; then the system analyzes the case similarities by coupling the CBR technique with a Naïve Bayes learner for feature-weights learning; and finally it produces an outcome recommendation. Our experimental results reveal that the CBR method with the implementation of a learner is indeed a viable alternative method to the forensic methods with practical advantages.     

Law, learning and representation

In machine learning terms, reasoning in legal cases can be compared to a lazy learning approach in which courts defer deciding how to generalize beyond the prior cases until the facts of a new case are observed. The HYPO family of systems implements a “lazy” approach since they defer making arguments how to decide a problem until the programs have positioned a new problem with respect to similar past cases. In a kind of “reflective adjustment”, they fit the new problem into a patchwork of past case decisions, comparing cases in order to reason about the legal significance of the relevant similarities and differences. Empirical evidence from diverse experiments shows that for purposes of teaching legal argumentation and performing legal information retrieval, HYPO-style systems' lazy learning approach and implementation of aspects of reflective adjustment can be very effective.