This book presents a methodology and philosophy of empirical science based on large scale lossless data compression. In this view a theory is scientific if it can be used to build a data compression program, and it is valuable if it can compress a standard benchmark database to a small size, taking into account the length of the compressor itself. This methodology therefore includes an Occam principle as well as a solution to the problem of demarcation. Because of the fundamental difficulty of lossless compression, this type of research must be empirical in nature: compression can only be achieved by discovering and characterizing empirical regularities in the data. Because of this, the philosophy provides a way to reformulate fields such as computer vision and computational linguistics as empirical sciences: the former by attempting to compress databases of natural images, the latter by attempting to compress large text databases. The book argues that the rigor and objectivity of the compression principle should set the stage for systematic progress in these fields. The argument is especially strong in the context of computer vision, which is plagued by chronic problems of evaluation. The book also considers the field of machine learning. Here the traditional approach requires that the models proposed to solve learning problems be extremely simple, in order to avoid overfitting. However, the world may contain intrinsically complex phenomena, which would require complex models to understand. The compression philosophy can justify complex models because of the large quantity of data being modeled (if the target database is 100 Gb, it is easy to justify a 10 Mb model). The complex models and abstractions learned on the basis of the raw data (images, language, etc) can then be reused to solve any specific learning problem, such as face recognition or machine translation.

Previous studies have demonstrated that encoding a Bayesian network into a SAT formula and then performing weighted model counting using a backtracking search algorithm can be an effective method for exact inference. In this paper, we present techniques for improving this approach for Bayesian networks with noisy-OR and noisy-MAX relations-- two relations that are widely used in practice as they can dramatically reduce the number of probabilities one needs to specify. In particular, we present two SAT encodings for noisy-OR and two encodings for noisy-MAX that exploit the structure or semantics of the relations to improve both time and space efficiency, and we prove the correctness of the encodings. We experimentally evaluated our techniques on large-scale real and randomly generated Bayesian networks. On these benchmarks, our techniques gave speedups of up to two orders of magnitude over the best previous approaches for networks with noisy-OR/MAX relations and scaled up to larger networks. As well, our techniques extend the weighted model counting approach for exact inference to networks that were previously intractable for the approach.

Pattern learning in an important problem in Natural Language Processing (NLP). Some exhaustive pattern learning (EPL) methods (Bod, 1992) were proved to be flawed (Johnson, 2002), while similar algorithms (Och and Ney, 2004) showed great advantages on other tasks, such as machine translation. In this article, we first formalize EPL, and then show that the probability given by an EPL model is constant-factor approximation of the probability given by an ensemble method that integrates exponential number of models obtained with various segmentations of the training data. This work for the first time provides theoretical justification for the widely used EPL algorithm in NLP, which was previously viewed as a flawed heuristic method. Better understanding of EPL may lead to improved pattern learning algorithms in future.

Sequential decision tasks present many opportunities for the study of transfer learning. A principal one among them is the existence of multiple domains that share the same underlying causal structure for actions. We describe an approach that exploits this shared causal structure to discover a hierarchical task structure in a source domain, which in turn speeds up learning of task execution knowledge in a new target domain. Our approach is theoretically justified and compares favorably to manually designed task hierarchies in learning efficiency in the target domain. We demonstrate that causally motivated task hierarchies transfer more robustly than other kinds of detailed knowledge that depend on the idiosyncrasies of the source domain and are hence less transferable.

Modern Internet services, such as those at Google, Yahoo!, and Amazon, handle billions of requests per day on clusters of thousands of computers. Because these services operate under strict performance requirements, a statistical understanding of their performance is of great practical interest. Such services are modeled by networks of queues, where each queue models one of the computers in the system. A key challenge is that the data are incomplete, because recording detailed information about every request to a heavily used system can require unacceptable overhead. In this paper we develop a Bayesian perspective on queueing models in which the arrival and departure times that are not observed are treated as latent variables. Underlying this viewpoint is the observation that a queueing model defines a deterministic transformation between the data and a set of independent variables called the service times. With this viewpoint in hand, we sample from the posterior distribution over missing data and model parameters using Markov chain Monte Carlo. We evaluate our framework on data from a benchmark Web application. We also present a simple technique for selection among nested queueing models. We are unaware of any previous work that considers inference in networks of queues in the presence of missing data.

Structural equation models and Bayesian networks have been widely used to analyze causal relations between continuous variables. In such frameworks, linear acyclic models are typically used to model the data-generating process of variables. Recently, it was shown that use of non-Gaussianity identifies the full structure of a linear acyclic model, i.e., a causal ordering of variables and their connection strengths, without using any prior knowledge on the network structure, which is not the case with conventional methods. However, existing estimation methods are based on iterative search algorithms and may not converge to a correct solution in a finite number of steps. In this paper, we propose a new direct method to estimate a causal ordering and connection strengths based on non-Gaussianity. In contrast to the previous methods, our algorithm requires no algorithmic parameters and is guaranteed to converge to the right solution within a small fixed number of steps if the data strictly follows the model.

To maximize its success, an AGI typically needs to explore its initially unknown world. Is there an optimal way of doing so? Here we derive an affirmative answer for a broad class of environments.

The challenges of effective health risk communication are well known. This paper provides pointers to the health communication literature that discuss these problems. Tailoring printed information, visual displays, and interactive multimedia have been proposed in the health communication literature as promising approaches. On-line risk communication applications are increasing on the internet. However, potential effectiveness of applications using conventional computer technology is limited. We propose that use of artificial intelligence, building upon research in Intelligent Tutoring Systems, might be able to overcome these limitations.

One challenge of building and maintaining large-scale data management systems is managing data fusion from multiple data sources. Often times, different data sources may represent the same data element in a slightly different way. These differences may represent an error in the data or a disagreement between sources on the correct value that best represents the data point. When the quantity of data managed and fused becomes sufficiently large, manual review becomes impossible, and automated systems must be built to manage data fusion. Some of the traditional solutions use simple voting theory, Dempster-Shafer theory, fuzzy matching and incremental learning. This paper presents a novel approach to data fusion in the domain of business listings. The task at hand, business listing categorization, suffers from conflicting and incomplete data from disparate data sources. Given the need for a high degree of accuracy in this task, we use a combination of case-based reasoning, joint probability, and domain-specific rules to improve data accuracy above other methods.

We introduce CIM, a Commonsense Inference Memory system utilizing both Extended Semantic Networks and Bayesian Networks that builds upon the commonsense knowledgebase ConceptNet. CIM introduces a new technique for self-assembling Bayesian Networks that allows only relevant parts of the commonsense database to affect the inference. The Bayesian Network include the activity in the input sentences and the related activities appearing in the commonsense database. They are used to interpret and infer the meaning of the set of sentences input. Without self-assembled networks, only relevant inference is performed, speeding up performance of reasoning with commonsense knowledge. We demonstrate that our system can disambiguate the needs of the user even if they do not state them directly, and do not use keywords. This ability would not be possible without either the use of commonsense or significant training. Eventually this approach may be applied to increase the effectiveness of other natural language understanding techniques as well.