The Association for the Advancement of Artificial Intelligence, in cooperation, with Stanford University's Department of Com-puter Science, presented the 1999 Spring Symposium Series on 22 to 24 March 1999 at Stanford University. The titles of the seven symposia were (1) Agents with Adjustable Autonomy, (2) Artificial Intelligence and Computer Games, (3) Artificial Intelligence in Equipment Maintenance Service and Support, (4) Hybrid Systems and AI: Modeling, Analysis, and Control of Discrete + Continuous Systems, (5) Intelligent Agents in Cyberspace, (6) Predictive Toxicology of Chemicals: Experiences and Impact of AI Tools, and (7) Search Techniques for Problem Solving under Uncertainty and Incomplete Information.

Intelligent-agent technology is one of the most exciting, active areas of research and development in computer science and information technology today. The First Asia-Pacific Conference on Intelligent- Agent Technology (IAT'99) attracted researchers and practitioners from diverse fields such as computer science, information systems, business, telecommunications, manufacturing, human factors, psychology, education, and robotics to examine the design principles and performance characteristics of various approaches in agent technologies and, hence, fostered the cross-fertilization of ideas on the development of autonomous agents and multiagent systems among different domains.

The 2000 Association of Computing Machinery Conference on Intelligent User Interfaces (IUI -- 2000) was held in New Orleans, Louisiana, from 9-12 January. This conference occupies the currently hot area that lies midway between the traditional fields of AI and computer-human interaction (CHI). For AI practitioners, this conference represents a good venue for learning about both how to design user interfaces for AI applications and how to use AI techniques to improve the user experience with more conventional applications. This year's conference drew the largest audience yet for an IUI conference, but the conference still remains at a manageable, single-track size. A wide range of high-quality presentations, tutorials, demonstrations, and invited speakers provided a bridge between the AI and CHI communities.

In 1998, the international planning community was invited to take part in the first planning competition, hosted by the Artificial Intelligence Planning Systems Conference, to provide a new impetus for empirical evaluation and direct comparison of automatic domain-independent planning systems. This article describes the systems that competed in the event, examines the results, and considers some of the implications for the future of the field.

I report on my experience over the past few years in introducing automated, model-based diagnostic technologies into industrial settings. In partic-ular, I discuss the competition that this technology has been receiving from handcrafted, rule-based diagnostic systems that has set some high standards that must be met by model-based systems before they can be viewed as viable alternatives. The battle between model-based and rule-based approaches to diagnosis has been over in the academic literature for many years, but the situation is different in industry where rule-based systems are dominant and appear to be attractive given the considerations of efficiency, embeddability, and cost effectiveness. My goal in this article is to provide a perspective on this competition and discuss a diagnostic tool, called DTOOL/CNETS, that I have been developing over the years as I tried to address the major challenges posed by rule-based systems. In particular, I discuss three major features of the developed tool that were either adopted, designed, or innovated to address these challenges: (1) its compositional modeling approach, (2) its structure-based computational approach, and (3) its ability to synthesize embeddable diagnostic systems for a variety of software and hardware platforms.

The 1998 Planning Competition at the AI Planning Systems Conference was the first of its kind. Its goal was to create planning domains that a wide variety of planning researchers could agree on to make comparison among planners more meaningful, measure overall progress in the field, and set up a framework for long-term creation of a repository of problems in a standard notation. A rules committee for the competition was created in 1997 and had long discussions on how the contest should go. One result of these discussions was the pddl notation for planning domains. This notation was used to set up a set of planning problems and get a modest problem repository started. As a result, five planning systems were able to compete when the contest took place in June 1998. All these systems solved problems in the strips framework, with some slight extensions. The attempt to find domains for other forms of planning foundered because of technical and organizational problems. In spite of this problem, the competition achieved its goals partially in that it con-firmed that substantial progress had occurred in some subfields of planning, and it allowed qualitative comparison among different planning algorithms. It is urged that the competition continue to take place and to evolve.

AAAI is pleased to present the elected fellows for 2000: Kenneth M. Ford, Eric Grimson, Leslie Pack Kaelbling, David Poole, Jonathan Schaeffer, and Bart Selman

This paper describes a novel method by which a spoken dialogue system can learn to choose an optimal dialogue strategy from its experience interacting with human users. The method is based on a combination of reinforcement learning and performance modeling of spoken dialogue systems. The reinforcement learning component applies Q-learning (Watkins, 1989), while the performance modeling component applies the PARADISE evaluation framework (Walker et al., 1997) to learn the performance function (reward) used in reinforcement learning. We illustrate the method with a spoken dialogue system named ELVIS (EmaiL Voice Interactive System), that supports access to email over the phone. We conduct a set of experiments for training an optimal dialogue strategy on a corpus of 219 dialogues in which human users interact with ELVIS over the phone. We then test that strategy on a corpus of 18 dialogues. We show that ELVIS can learn to optimize its strategy selection for agent initiative, for reading messages, and for summarizing email folders.

The paper addresses the problem of computing goal orderings, which is one of the longstanding issues in AI planning. It makes two new contributions. First, it formally defines and discusses two different goal orderings, which are called the reasonable and the forced ordering. Both orderings are defined for simple STRIPS operators as well as for more complex ADL operators supporting negation and conditional effects. The complexity of these orderings is investigated and their practical relevance is discussed. Secondly, two different methods to compute reasonable goal orderings are developed. One of them is based on planning graphs, while the other investigates the set of actions directly. Finally, it is shown how the ordering relations, which have been derived for a given set of goals G, can be used to compute a so-called goal agenda that divides G into an ordered set of subgoals. Any planner can then, in principle, use the goal agenda to plan for increasing sets of subgoals. This can lead to an exponential complexity reduction, as the solution to a complex planning problem is found by solving easier subproblems. Since only a polynomial overhead is caused by the goal agenda computation, a potential exists to dramatically speed up planning algorithms as we demonstrate in the empirical evaluation, where we use this method in the IPP planner.

The recent approaches of extending the GRAPHPLAN algorithm to handle more expressive planning formalisms raise the question of what the formal meaning of ``expressive power'' is. We formalize the intuition that expressive power is a measure of how concisely planning domains and plans can be expressed in a particular formalism by introducing the notion of ``compilation schemes'' between planning formalisms. Using this notion, we analyze the expressiveness of a large family of propositional planning formalisms, ranging from basic STRIPS to a formalism with conditional effects, partial state specifications, and propositional formulae in the preconditions. One of the results is that conditional effects cannot be compiled away if plan size should grow only linearly but can be compiled away if we allow for polynomial growth of the resulting plans. This result confirms that the recently proposed extensions to the GRAPHPLAN algorithm concerning conditional effects are optimal with respect to the ``compilability'' framework. Another result is that general propositional formulae cannot be compiled into conditional effects if the plan size should be preserved linearly. This implies that allowing general propositional formulae in preconditions and effect conditions adds another level of difficulty in generating a plan.