Goto

Collaborating Authors

 Genre


Automated Deduction: Looking Ahead

AI Magazine

To not only proving new mathematical obtain broader input, especially from countries results by computer but also formally verifying outside North America, a call for commentaries the correctness of (certain properties of) computer was issued to the automated deduction community.


Turbine Engine Diagnostics (TED)

AI Magazine

Turbine engine diagnostics (TED) is a diagnostic expert system to aid the M1 Abrams tank mechanic find-and-fix problems in the AGT-1500 turbine engine. TED was designed to provide the apprentice mechanic with the ability to diagnose and repair the turbine engine like an expert mechanic. The expert system was designed and built by the U.S. Army Research Laboratory and the U.S. Army Ordnance Center and School. This article discusses the relevant background, development issues, reasoning method, system overview, test results, return on investment, and fielding history of the project. Limited fielding began in 1994 to select U.S. Army National Guard units and complete fielding to all M1 Abrams tank maintenance units started in 1997 and will finish by the end of 1998. The Army estimates that TED will save roughly $10 million a year through improved diagnostic accuracy and reduced waste. The development and fielding of the TED program represents the Army's first successful fielded maintenance system in the area of AI. Several reasons can be given for the success of the TED program: an appropriate domain with proper scope, a close relationship with the expert, extensive user involvement, and others that are discussed in this article.


AI Game-Playing Techniques

AI Magazine

In conjunction with the Association for the Advancement of Artificial Intelligence's Hall of Champions exhibit, the Innovative Applications of Artificial Intelligence held a panel discussion entitled "AI Game-Playing Techniques: Are They Useful for Anything Other Than Games?" This article summarizes the panelists' comments about whether ideas and techniques from AI game playing are useful elsewhere and what kinds of game might be suitable as "challenge problems" for future research.


Minimum Description Length Induction, Bayesianism, and Kolmogorov Complexity

arXiv.org Artificial Intelligence

The relationship between the Bayesian approach and the minimum description length approach is established. We sharpen and clarify the general modeling principles MDL and MML, abstracted as the ideal MDL principle and defined from Bayes's rule by means of Kolmogorov complexity. The basic condition under which the ideal principle should be applied is encapsulated as the Fundamental Inequality, which in broad terms states that the principle is valid when the data are random, relative to every contemplated hypothesis and also these hypotheses are random relative to the (universal) prior. Basically, the ideal principle states that the prior probability associated with the hypothesis should be given by the algorithmic universal probability, and the sum of the log universal probability of the model plus the log of the probability of the data given the model should be minimized. If we restrict the model class to the finite sets then application of the ideal principle turns into Kolmogorov's minimal sufficient statistic. In general we show that data compression is almost always the best strategy, both in hypothesis identification and prediction.


TDLeaf(lambda): Combining Temporal Difference Learning with Game-Tree Search

arXiv.org Artificial Intelligence

In this paper we present TDLeaf(lambda), a variation on the TD(lambda) algorithm that enables it to be used in conjunction with minimax search. We present some experiments in both chess and backgammon which demonstrate its utility and provide comparisons with TD(lambda) and another less radical variant, TD-directed(lambda). In particular, our chess program, ``KnightCap,'' used TDLeaf(lambda) to learn its evaluation function while playing on the Free Internet Chess Server (FICS, fics.onenet.net). It improved from a 1650 rating to a 2100 rating in just 308 games. We discuss some of the reasons for this success and the relationship between our results and Tesauro's results in backgammon.


Experiences with Bayesian Learning in a Real World Application

Neural Information Processing Systems

This paper reports about an application of Bayes' inferred neural network classifiers in the field of automatic sleep staging. The reason for using Bayesian learning for this task is twofold. First, Bayesian inference is known to embody regularization automatically. Second, a side effect of Bayesian learning leads to larger variance of network outputs in regions without training data. This results in well known moderation effects, which can be used to detect outliers. In a 5 fold cross-validation experiment the full Bayesian solution found with R. Neals hybrid Monte Carlo algorithm, was not better than a single maximum a-posteriori (MAP) solution found with D.J. MacKay's evidence approximation. In a second experiment we studied the properties of both solutions in rejecting classification of movement artefacts.


Adaptation in Speech Motor Control

Neural Information Processing Systems

Human subjects are known to adapt their motor behavior to a shift of the visual field brought about by wearing prism glasses over their eyes. We have studied the analog of this effect in speech. U sing a device that can feed back transformed speech signals in real time, we exposed subjects to alterations of their own speech feedback. We found that speakers learn to adjust their production of a vowel to compensate for feedback alterations that change the vowel's perceived phonetic identity; moreover, the effect generalizes across consonant contexts and to different vowels. 1 INTRODUCTION For more than a century, it has been know that humans will adapt their reaches to altered visual feedback [8]. One of the most studied examples of this adaptation is prism adaptation, which is seen when a subject reaches to targets while wearing image-shifting prism glasses [2].


Bayesian Robustification for Audio Visual Fusion

Neural Information Processing Systems

Department of Cognitive Science Department of Cognitive Science University of California, San Diego University of California, San Diego La Jolla, CA 92092-0515 La Jolla, CA 92092-0515 Abstract We discuss the problem of catastrophic fusion in multimodal recognition systems. This problem arises in systems that need to fuse different channels in non-stationary environments. Practice shows that when recognition modules within each modality are tested in contexts inconsistent with their assumptions, their influence on the fused product tends to increase, with catastrophic results. We explore a principled solution to this problem based upon Bayesian ideas of competitive models and inference robustification: each sensory channel is provided with simple white-noise context models, and the perceptual hypothesis and context are jointly estimated. Consequently, context deviations are interpreted as changes in white noise contamination strength, automatically adjusting the influence of the module.


Approximating Posterior Distributions in Belief Networks Using Mixtures

Neural Information Processing Systems

Exact inference in densely connected Bayesian networks is computationally intractable, and so there is considerable interest in developing effective approximation schemes. One approach which has been adopted is to bound the log likelihood using a mean-field approximating distribution. While this leads to a tractable algorithm, the mean field distribution is assumed to be factorial and hence unimodal. In this paper we demonstrate the feasibility of using a richer class of approximating distributions based on mixtures of mean field distributions. We derive an efficient algorithm for updating the mixture parameters and apply it to the problem of learning in sigmoid belief networks. Our results demonstrate a systematic improvement over simple mean field theory as the number of mixture components is increased.


Linear Concepts and Hidden Variables: An Empirical Study

Neural Information Processing Systems

Some learning techniques for classification tasks work indirectly, by first trying to fit a full probabilistic model to the observed data. Whether this is a good idea or not depends on the robustness with respect to deviations from the postulated model. We study this question experimentally in a restricted, yet nontrivial and interesting case: we consider a conditionally independent attribute (CIA) model which postulates a single binary-valued hidden variable z on which all other attributes (i.e., the target and the observables) depend. In this model, finding the most likely value of anyone variable (given known values for the others) reduces to testing a linear function of the observed values. We learn CIA with two techniques: the standard EM algorithm, and a new algorithm we develop based on covariances. We compare these, in a controlled fashion, against an algorithm (a version of Winnow) that attempts to find a good linear classifier directly. Our conclusions help delimit the fragility of using the CIA model for classification: once the data departs from this model, performance quickly degrades and drops below that of the directly-learned linear classifier.