This paper presents an approach to learning from situated, interactive tutorial instruction within an ongoing agent. Tutorial instruction is a flexible (and thus powerful) paradigm for teaching tasks because it allows an instructor to communicate whatever types of knowledge an agent might need in whatever situations might arise. To support this flexibility, however, the agent must be able to learn multiple kinds of knowledge from a broad range of instructional interactions. Our approach, called situated explanation, achieves such learning through a combination of analytic and inductive techniques. It combines a form of explanation-based learning that is situated for each instruction with a full suite of contextually guided responses to incomplete explanations. The approach is implemented in an agent called Instructo-Soar that learns hierarchies of new tasks and other domain knowledge from interactive natural language instructions. Instructo-Soar meets three key requirements of flexible...

In contrast to current intelligent systems, which must be laboriously programmed for each task they are meant to perform, instructable agents can be taught new tasks and associated knowledge. This thesis presents a general theory of learning from tutorial instruction and its use to produce an instructable agent. Tutorial instruction is a particularly powerful form of instruction, because it allows the instructor to communicate whatever kind of knowledge a student needs at whatever point it is needed. To exploit this broad flexibility, however, a tutorable agent must support a full range of interaction with its instructor to learn a full range of knowledge. Thus, unlike most machine learning tasks, which target deep learning of a single kind of knowledge from a single kind of input, tutorability requires a breadth of learning from a broad range of instructional interactions. The theory of learning from tutorial...

CONTENTS Page ACKNOWLEDGEMENT . . . . . . . . . . . . . . . . . . . . iii LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . vi CHAPTER I. INTRODUCTION . . . . . . . . . . . . . . . . 1 1.1 An Overview . . . . . . . . . . . . . . . 1 1.2 Evolution of Computer-Based Instruction at Rush . . . . . . . . . . . . . . . . . 3 1.3 Goals of the Thesis . . . . . . . . . . . 4 1.4 Organization of the Thesis . . . . . . . . 6 II. THE BACKGROUND . . . . . . . . . . . . . . . 9 2.1 Qualitative Reasoning . . . . . . . . . . 9 2.2 Subject Area . . . . . . . . . . . . . . 10 2.3 Organization . . . . . . . . . . . . . . 12 2.4 System Constraints . . . . . . . . . . . 14 2.5 Multiple Simultaneous Inputs . . . . . . . 15 III. ORGANIZATION OF CIRCSIM-TUTOR . . . . . . . . 18 3.1 Intelligent Tutoring Systems . . . . . . . 18 3.2 Domain Expertise . . . . . . . . . . . . 23 3.3 Input-Understander . . . . . . . . . . . 26 3.4 Student Modeler . . . . . . . . . . . . . 27 3.5 Instructional Planner . . . .

Several theories have been proposed regarding how causal relations among features of objects affect how those objects are classified. The assumptions of these theories were tested in 3 experiments that manipulated the causal knowledge associated with novel categories. There were 3 results. The 1st was a multiple cause effect in which a feature’s importance increases with its number of causes. The 2nd was a coherence effect in which good category members are those whose features jointly corroborate the category’s causal knowledge. These 2 effects can be accounted for by assuming that good category members are those likely to be generated by a category’s causal laws. The 3rd result was a primary cause effect, in which primary causes are more important to category membership. This effect can also be explained by a generative account with an additional assumption: that categories often are perceived to have hidden generative causes.

We describe an incremental learning algorithm, called theory-driven learning, that creates rules to predict the effect of actions. Theory-driven learning exploits knowledge of regularities among rules to constrain learning. We demonstrate that this knowledge enables the learning system to rapidly converge on accurate predictive rules and to tolerate more complex training data. An algorithm for incrementally learning these regularities is described and we provide evidence that the resulting regularities are sufficiently general to facilitate learning in new domains. The results demonstrate transfer from one domain to another can be achieved by deliberately overgeneralizing rules in one domain and biasing the learning algorithm to create new rules that specialize these overgeneralizations in other domains.

We present a memory-based approach to learning commonsense causal relations from episodic text. The method relies on dynamic memory which consists of events, event schemata, episodes, causal heuristics, and causal hypotheses. The learning algorithms are based on applying causal heuristics to precedents of new information. The heuristics are derived from principles of causation, and, to a limited extent, from domain-related causal reasoning. Learning is defined as finding---and later augmenting---inter-episodal and intra-episodal causal connections. The learning algorithms enable inductive generalization of causal associations into AND/OR graphs. The methodology has been implemented and tested in the program NEXUS. Memory-based hypothesis Error! Unknown switch argument. INTRODUCTION In this paper, we examine the mechanisms by which causal relations expressed in natural language can be learned. Natural languages provide ample means to describe physical and mental events, marked relati...

Causal learning requires integrating constraints provided by domain-specific theories with domaingeneral statistical learning. In order to investigate the interaction between these factors, the authors presented preschoolers with stories pitting their existing theories against statistical evidence. Each child heard 2 stories in which 2 candidate causes co-occurred with an effect. Evidence was presented in the

Co-Principal Investigators. Any opinions, findings, conclusions or recommendations expressed here are those of the authors and do not necessarily reflect the views of the National Science Foundation. Models and Moves Extensive research on students ’ understanding of science has documented persistent shortfalls at all ages. One way to account for students ’ difficulties is to consider the particular challenges posed by individual science concepts. This article offers an alternative view. We argue that students’ difficulties in large part reflect unfamiliarity with a small number of causal modeling styles charac-teristic of received scientific models. These include, for instance, explaining surface phenomena with an underlying mechanism, relying on constraint-system explanations as in Ohm’s law, includ-ing probabilistic elements as in chaos theory, and acknowledging causal webs and self-organizing systems as in ecologies--in sum, aspects of “complex causality. ” A further barrier to understanding is students ’ unfamiliarity with epistemic moves that might challenge their initial explanations, such as looking for missing links in a causal story or putting a model at risk. The article offers a frame-work for classifying aspects of complex causality in modeling and for supporting epistemic moves. Empirical research both from the literature and our own work is presented in support of the framework, including evidence that instruction based on causal models and epistemic moves enhances students ’ understanding of science concepts. 2

This paper is based on the results of research carried out during the first year and a half of the Understandings of Consequence Project. We are continuing to research and develop the ideas presented here. If you have feedback for us or would like to keep in touch with developments on the project, please check our website at http://pzweb.harvard.edu/Research/UnderCon.htm or send us an email at Tina_Grotzer @PZ.Harvard.Edu.