id
stringlengths
9
16
title
stringlengths
4
278
categories
stringlengths
5
104
abstract
stringlengths
6
4.09k
cs/0209021
Activities, Context and Ubiquitous Computing
cs.IR
Context and context-awareness provides computing environments with the ability to usefully adapt the services or information they provide. It is the ability to implicitly sense and automatically derive the user needs that separates context-aware applications from traditionally designed applications, and this makes them more attentive, responsive, and aware of their user's identity, and their user's environment. This paper argues that context-aware applications capable of supporting complex, cognitive activities can be built from a model of context called Activity-Centric Context. A conceptual model of Activity-Centric context is presented. The model is illustrated via a detailed example.
cs/0209022
A Comparison of Different Cognitive Paradigms Using Simple Animats in a Virtual Laboratory, with Implications to the Notion of Cognition
cs.AI
In this thesis I present a virtual laboratory which implements five different models for controlling animats: a rule-based system, a behaviour-based system, a concept-based system, a neural network, and a Braitenberg architecture. Through different experiments, I compare the performance of the models and conclude that there is no "best" model, since different models are better for different things in different contexts. The models I chose, although quite simple, represent different approaches for studying cognition. Using the results as an empirical philosophical aid, I note that there is no "best" approach for studying cognition, since different approaches have all advantages and disadvantages, because they study different aspects of cognition from different contexts. This has implications for current debates on "proper" approaches for cognition: all approaches are a bit proper, but none will be "proper enough". I draw remarks on the notion of cognition abstracting from all the approaches used to study it, and propose a simple classification for different types of cognition.
cs/0209030
Extremal Optimization: an Evolutionary Local-Search Algorithm
cs.NE cs.AI
A recently introduced general-purpose heuristic for finding high-quality solutions for many hard optimization problems is reviewed. The method is inspired by recent progress in understanding far-from-equilibrium phenomena in terms of {\em self-organized criticality,} a concept introduced to describe emergent complexity in physical systems. This method, called {\em extremal optimization,} successively replaces the value of extremely undesirable variables in a sub-optimal solution with new, random ones. Large, avalanche-like fluctuations in the cost function self-organize from this dynamics, effectively scaling barriers to explore local optima in distant neighborhoods of the configuration space while eliminating the need to tune parameters. Drawing upon models used to simulate the dynamics of granular media, evolution, or geology, extremal optimization complements approximation methods inspired by equilibrium statistical physics, such as {\em simulated annealing}. It may be but one example of applying new insights into {\em non-equilibrium phenomena} systematically to hard optimization problems. This method is widely applicable and so far has proved competitive with -- and even superior to -- more elaborate general-purpose heuristics on testbeds of constrained optimization problems with up to $10^5$ variables, such as bipartitioning, coloring, and satisfiability. Analysis of a suitable model predicts the only free parameter of the method in accordance with all experimental results.
cs/0210004
Revising Partially Ordered Beliefs
cs.AI
This paper deals with the revision of partially ordered beliefs. It proposes a semantic representation of epistemic states by partial pre-orders on interpretations and a syntactic representation by partially ordered belief bases. Two revision operations, the revision stemming from the history of observations and the possibilistic revision, defined when the epistemic state is represented by a total pre-order, are generalized, at a semantic level, to the case of a partial pre-order on interpretations, and at a syntactic level, to the case of a partially ordered belief base. The equivalence between the two representations is shown for the two revision operations.
cs/0210005
Positive time fractional derivative
cs.CE
In mathematical modeling of the non-squared frequency-dependent diffusions, also known as the anomalous diffusions, it is desirable to have a positive real Fourier transform for the time derivative of arbitrary fractional or odd integer order. The Fourier transform of the fractional time derivative in the Riemann-Liouville and Caputo senses, however, involves a complex power function of the fractional order. In this study, a positive time derivative of fractional or odd integer order is introduced to respect the positivity in modeling the anomalous diffusions.
cs/0210007
Compilability of Abduction
cs.AI cs.CC
Abduction is one of the most important forms of reasoning; it has been successfully applied to several practical problems such as diagnosis. In this paper we investigate whether the computational complexity of abduction can be reduced by an appropriate use of preprocessing. This is motivated by the fact that part of the data of the problem (namely, the set of all possible assumptions and the theory relating assumptions and manifestations) are often known before the rest of the problem. In this paper, we show some complexity results about abduction when compilation is allowed.
cs/0210009
On the Cell-based Complexity of Recognition of Bounded Configurations by Finite Dynamic Cellular Automata
cs.CC cs.CV
This paper studies complexity of recognition of classes of bounded configurations by a generalization of conventional cellular automata (CA) -- finite dynamic cellular automata (FDCA). Inspired by the CA-based models of biological and computer vision, this study attempts to derive the properties of a complexity measure and of the classes of input configurations that make it beneficial to realize the recognition via a two-layered automaton as compared to a one-layered automaton. A formalized model of an image pattern recognition task is utilized to demonstrate that the derived conditions can be satisfied for a non-empty set of practical problems.
cs/0210012
Selection of future events from a time series in relation to estimations of forecasting uncertainty
cs.NE
A new general procedure for a priori selection of more predictable events from a time series of observed variable is proposed. The procedure is applicable to time series which contains different types of events that feature significantly different predictability, or, in other words, to heteroskedastic time series. A priori selection of future events in accordance to expected uncertainty of their forecasts may be helpful for making practical decisions. The procedure first implies creation of two neural network based forecasting models, one of which is aimed at prediction of conditional mean and other - conditional dispersion, and then elaboration of the rule for future event selection into groups of more and less predictable events. The method is demonstrated and tested by the example of the computer generated time series, and then applied to the real world time series, Dow Jones Industrial Average index.
cs/0210018
User software for the next generation
cs.GR cs.CE
New generations of neutron scattering sources and instrumentation are providing challenges in data handling for user software. Time-of-Flight instruments used at pulsed sources typically produce hundreds or thousands of channels of data for each detector segment. New instruments are being designed with thousands to hundreds of thousands of detector segments. High intensity neutron sources make possible parametric studies and texture studies which further increase data handling requirements. The Integrated Spectral Analysis Workbench (ISAW) software developed at Argonne handles large numbers of spectra simultaneously while providing operations to reduce, sort, combine and export the data. It includes viewers to inspect the data in detail in real time. ISAW uses existing software components and packages where feasible and takes advantage of the excellent support for user interface design and network communication in Java. The included scripting language simplifies repetitive operations for analyzing many files related to a given experiment. Recent additions to ISAW include a contour view, a time-slice table view, routines for finding and fitting peaks in data, and support for data from other facilities using the NeXus format. In this paper, I give an overview of features and planned improvements of ISAW. Details of some of the improvements are covered in other presentations at this conference.
cs/0210023
Geometric Aspects of Multiagent Systems
cs.MA cs.AI
Recent advances in Multiagent Systems (MAS) and Epistemic Logic within Distributed Systems Theory, have used various combinatorial structures that model both the geometry of the systems and the Kripke model structure of models for the logic. Examining one of the simpler versions of these models, interpreted systems, and the related Kripke semantics of the logic $S5_n$ (an epistemic logic with $n$-agents), the similarities with the geometric / homotopy theoretic structure of groupoid atlases is striking. These latter objects arise in problems within algebraic K-theory, an area of algebra linked to the study of decomposition and normal form theorems in linear algebra. They have a natural well structured notion of path and constructions of path objects, etc., that yield a rich homotopy theory.
cs/0210025
An Algorithm for Pattern Discovery in Time Series
cs.LG cs.CL
We present a new algorithm for discovering patterns in time series and other sequential data. We exhibit a reliable procedure for building the minimal set of hidden, Markovian states that is statistically capable of producing the behavior exhibited in the data -- the underlying process's causal states. Unlike conventional methods for fitting hidden Markov models (HMMs) to data, our algorithm makes no assumptions about the process's causal architecture (the number of hidden states and their transition structure), but rather infers it from the data. It starts with assumptions of minimal structure and introduces complexity only when the data demand it. Moreover, the causal states it infers have important predictive optimality properties that conventional HMM states lack. We introduce the algorithm, review the theory behind it, prove its asymptotic reliability, use large deviation theory to estimate its rate of convergence, and compare it to other algorithms which also construct HMMs from data. We also illustrate its behavior on an example process, and report selected numerical results from an implementation.
cs/0210026
Encoding a Taxonomy of Web Attacks with Different-Length Vectors
cs.CR cs.AI
Web attacks, i.e. attacks exclusively using the HTTP protocol, are rapidly becoming one of the fundamental threats for information systems connected to the Internet. When the attacks suffered by web servers through the years are analyzed, it is observed that most of them are very similar, using a reduced number of attacking techniques. It is generally agreed that classification can help designers and programmers to better understand attacks and build more secure applications. As an effort in this direction, a new taxonomy of web attacks is proposed in this paper, with the objective of obtaining a practically useful reference framework for security applications. The use of the taxonomy is illustrated by means of multiplatform real world web attack examples. Along with this taxonomy, important features of each attack category are discussed. A suitable semantic-dependent web attack encoding scheme is defined that uses different-length vectors. Possible applications are described, which might benefit from this taxonomy and encoding scheme, such as intrusion detection systems and application firewalls.
cs/0210027
A uniform approach to logic programming semantics
cs.AI cs.LO
Part of the theory of logic programming and nonmonotonic reasoning concerns the study of fixed-point semantics for these paradigms. Several different semantics have been proposed during the last two decades, and some have been more successful and acknowledged than others. The rationales behind those various semantics have been manifold, depending on one's point of view, which may be that of a programmer or inspired by commonsense reasoning, and consequently the constructions which lead to these semantics are technically very diverse, and the exact relationships between them have not yet been fully understood. In this paper, we present a conceptually new method, based on level mappings, which allows to provide uniform characterizations of different semantics for logic programs. We will display our approach by giving new and uniform characterizations of some of the major semantics, more particular of the least model semantics for definite programs, of the Fitting semantics, and of the well-founded semantics. A novel characterization of the weakly perfect model semantics will also be provided.
cs/0210028
Equivalences Among Aggregate Queries with Negation
cs.DB cs.LO
Query equivalence is investigated for disjunctive aggregate queries with negated subgoals, constants and comparisons. A full characterization of equivalence is given for the aggregation functions count, max, sum, prod, toptwo and parity. A related problem is that of determining, for a given natural number N, whether two given queries are equivalent over all databases with at most N constants. We call this problem bounded equivalence. A complete characterization of decidability of bounded equivalence is given. In particular, it is shown that this problem is decidable for all the above aggregation functions as well as for count distinct and average. For quasilinear queries (i.e., queries where predicates that occur positively are not repeated) it is shown that equivalence can be decided in polynomial time for the aggregation functions count, max, sum, parity, prod, toptwo and average. A similar result holds for count distinct provided that a few additional conditions hold. The results are couched in terms of abstract characteristics of aggregation functions, and new proof techniques are used. Finally, the results above also imply that equivalence, under bag-set semantics, is decidable for non-aggregate queries with negation.
cs/0210030
Intelligence and Cooperative Search by Coupled Local Minimizers
cs.AI cs.MA cs.NE
We show how coupling of local optimization processes can lead to better solutions than multi-start local optimization consisting of independent runs. This is achieved by minimizing the average energy cost of the ensemble, subject to synchronization constraints between the state vectors of the individual local minimizers. From an augmented Lagrangian which incorporates the synchronization constraints both as soft and hard constraints, a network is derived wherein the local minimizers interact and exchange information through the synchronization constraints. From the viewpoint of neural networks, the array can be considered as a Lagrange programming network for continuous optimization and as a cellular neural network (CNN). The penalty weights associated with the soft state synchronization constraints follow from the solution to a linear program. This expresses that the energy cost of the ensemble should maximally decrease. In this way successful local minimizers can implicitly impose their state to the others through a mechanism of master-slave dynamics resulting into a cooperative search mechanism. Improved information spreading within the ensemble is obtained by applying the concept of small-world networks. This work suggests, in an interdisciplinary context, the importance of information exchange and state synchronization within ensembles, towards issues as evolution, collective behaviour, optimality and intelligence.
cs/0211003
Evaluation of the Performance of the Markov Blanket Bayesian Classifier Algorithm
cs.LG
The Markov Blanket Bayesian Classifier is a recently-proposed algorithm for construction of probabilistic classifiers. This paper presents an empirical comparison of the MBBC algorithm with three other Bayesian classifiers: Naive Bayes, Tree-Augmented Naive Bayes and a general Bayesian network. All of these are implemented using the K2 framework of Cooper and Herskovits. The classifiers are compared in terms of their performance (using simple accuracy measures and ROC curves) and speed, on a range of standard benchmark data sets. It is concluded that MBBC is competitive in terms of speed and accuracy with the other algorithms considered.
cs/0211004
The DLV System for Knowledge Representation and Reasoning
cs.AI cs.LO cs.PL
This paper presents the DLV system, which is widely considered the state-of-the-art implementation of disjunctive logic programming, and addresses several aspects. As for problem solving, we provide a formal definition of its kernel language, function-free disjunctive logic programs (also known as disjunctive datalog), extended by weak constraints, which are a powerful tool to express optimization problems. We then illustrate the usage of DLV as a tool for knowledge representation and reasoning, describing a new declarative programming methodology which allows one to encode complex problems (up to $\Delta^P_3$-complete problems) in a declarative fashion. On the foundational side, we provide a detailed analysis of the computational complexity of the language of DLV, and by deriving new complexity results we chart a complete picture of the complexity of this language and important fragments thereof. Furthermore, we illustrate the general architecture of the DLV system which has been influenced by these results. As for applications, we overview application front-ends which have been developed on top of DLV to solve specific knowledge representation tasks, and we briefly describe the main international projects investigating the potential of the system for industrial exploitation. Finally, we report about thorough experimentation and benchmarking, which has been carried out to assess the efficiency of the system. The experimental results confirm the solidity of DLV and highlight its potential for emerging application areas like knowledge management and information integration.
cs/0211005
Prosody Based Co-analysis for Continuous Recognition of Coverbal Gestures
cs.CV cs.HC
Although speech and gesture recognition has been studied extensively, all the successful attempts of combining them in the unified framework were semantically motivated, e.g., keyword-gesture cooccurrence. Such formulations inherited the complexity of natural language processing. This paper presents a Bayesian formulation that uses a phenomenon of gesture and speech articulation for improving accuracy of automatic recognition of continuous coverbal gestures. The prosodic features from the speech signal were coanalyzed with the visual signal to learn the prior probability of co-occurrence of the prominent spoken segments with the particular kinematical phases of gestures. It was found that the above co-analysis helps in detecting and disambiguating visually small gestures, which subsequently improves the rate of continuous gesture recognition. The efficacy of the proposed approach was demonstrated on a large database collected from the weather channel broadcast. This formulation opens new avenues for bottom-up frameworks of multimodal integration.
cs/0211006
Maximing the Margin in the Input Space
cs.AI cs.LG
We propose a novel criterion for support vector machine learning: maximizing the margin in the input space, not in the feature (Hilbert) space. This criterion is a discriminative version of the principal curve proposed by Hastie et al. The criterion is appropriate in particular when the input space is already a well-designed feature space with rather small dimensionality. The definition of the margin is generalized in order to represent prior knowledge. The derived algorithm consists of two alternating steps to estimate the dual parameters. Firstly, the parameters are initialized by the original SVM. Then one set of parameters is updated by Newton-like procedure, and the other set is updated by solving a quadratic programming problem. The algorithm converges in a few steps to a local optimum under mild conditions and it preserves the sparsity of support vectors. Although the complexity to calculate temporal variables increases the complexity to solve the quadratic programming problem for each step does not change. It is also shown that the original SVM can be seen as a special case. We further derive a simplified algorithm which enables us to use the existing code for the original SVM.
cs/0211007
Approximating Incomplete Kernel Matrices by the em Algorithm
cs.LG
In biological data, it is often the case that observed data are available only for a subset of samples. When a kernel matrix is derived from such data, we have to leave the entries for unavailable samples as missing. In this paper, we make use of a parametric model of kernel matrices, and estimate missing entries by fitting the model to existing entries. The parametric model is created as a set of spectral variants of a complete kernel matrix derived from another information source. For model fitting, we adopt the em algorithm based on the information geometry of positive definite matrices. We will report promising results on bacteria clustering experiments using two marker sequences: 16S and gyrB.
cs/0211008
Can the whole brain be simpler than its "parts"?
cs.AI
This is the first in a series of connected papers discussing the problem of a dynamically reconfigurable universal learning neurocomputer that could serve as a computational model for the whole human brain. The whole series is entitled "The Brain Zero Project. My Brain as a Dynamically Reconfigurable Universal Learning Neurocomputer." (For more information visit the website www.brain0.com.) This introductory paper is concerned with general methodology. Its main goal is to explain why it is critically important for both neural modeling and cognitive modeling to pay much attention to the basic requirements of the whole brain as a complex computing system. The author argues that it can be easier to develop an adequate computational model for the whole "unprogrammed" (untrained) human brain than to find adequate formal representations of some nontrivial parts of brain's performance. (In the same way as, for example, it is easier to describe the behavior of a complex analytical function than the behavior of its real and/or imaginary part.) The "curse of dimensionality" that plagues purely phenomenological ("brainless") cognitive theories is a natural penalty for an attempt to represent insufficiently large parts of brain's performance in a state space of insufficiently high dimensionality. A "partial" modeler encounters "Catch 22." An attempt to simplify a cognitive problem by artificially reducing its dimensionality makes the problem more difficult.
cs/0211014
Vanquishing the XCB Question: The Methodology Discovery of the Last Shortest Single Axiom for the Equivalential Calculus
cs.LO cs.AI
With the inclusion of an effective methodology, this article answers in detail a question that, for a quarter of a century, remained open despite intense study by various researchers. Is the formula XCB = e(x,e(e(e(x,y),e(z,y)),z)) a single axiom for the classical equivalential calculus when the rules of inference consist of detachment (modus ponens) and substitution? Where the function e represents equivalence, this calculus can be axiomatized quite naturally with the formulas e(x,x), e(e(x,y),e(y,x)), and e(e(x,y),e(e(y,z),e(x,z))), which correspond to reflexivity, symmetry, and transitivity, respectively. (We note that e(x,x) is dependent on the other two axioms.) Heretofore, thirteen shortest single axioms for classical equivalence of length eleven had been discovered, and XCB was the only remaining formula of that length whose status was undetermined. To show that XCB is indeed such a single axiom, we focus on the rule of condensed detachment, a rule that captures detachment together with an appropriately general, but restricted, form of substitution. The proof we present in this paper consists of twenty-five applications of condensed detachment, completing with the deduction of transitivity followed by a deduction of symmetry. We also discuss some factors that may explain in part why XCB resisted relinquishing its treasure for so long. Our approach relied on diverse strategies applied by the automated reasoning program OTTER. Thus ends the search for shortest single axioms for the equivalential calculus.
cs/0211015
XCB, the Last of the Shortest Single Axioms for the Classical Equivalential Calculus
cs.LO cs.AI
It has long been an open question whether the formula XCB = EpEEEpqErqr is, with the rules of substitution and detachment, a single axiom for the classical equivalential calculus. This paper answers that question affirmatively, thus completing a search for all such eleven-symbol single axioms that began seventy years ago.
cs/0211017
Probabilistic Parsing Strategies
cs.CL
We present new results on the relation between purely symbolic context-free parsing strategies and their probabilistic counter-parts. Such parsing strategies are seen as constructions of push-down devices from grammars. We show that preservation of probability distribution is possible under two conditions, viz. the correct-prefix property and the property of strong predictiveness. These results generalize existing results in the literature that were obtained by considering parsing strategies in isolation. From our general results we also derive negative results on so-called generalized LR parsing.
cs/0211020
Monadic Datalog and the Expressive Power of Languages for Web Information Extraction
cs.DB
Research on information extraction from Web pages (wrapping) has seen much activity recently (particularly systems implementations), but little work has been done on formally studying the expressiveness of the formalisms proposed or on the theoretical foundations of wrapping. In this paper, we first study monadic datalog over trees as a wrapping language. We show that this simple language is equivalent to monadic second order logic (MSO) in its ability to specify wrappers. We believe that MSO has the right expressiveness required for Web information extraction and propose MSO as a yardstick for evaluating and comparing wrappers. Along the way, several other results on the complexity of query evaluation and query containment for monadic datalog over trees are established, and a simple normal form for this language is presented. Using the above results, we subsequently study the kernel fragment Elog$^-$ of the Elog wrapping language used in the Lixto system (a visual wrapper generator). Curiously, Elog$^-$ exactly captures MSO, yet is easier to use. Indeed, programs in this language can be entirely visually specified.
cs/0211023
SkyQuery: A WebService Approach to Federate Databases
cs.DB cs.CE
Traditional science searched for new objects and phenomena that led to discoveries. Tomorrow's science will combine together the large pool of information in scientific archives and make discoveries. Scienthists are currently keen to federate together the existing scientific databases. The major challenge in building a federation of these autonomous and heterogeneous databases is system integration. Ineffective integration will result in defunct federations and under utilized scientific data. Astronomy, in particular, has many autonomous archives spread over the Internet. It is now seeking to federate these, with minimal effort, into a Virtual Observatory that will solve complex distributed computing tasks such as answering federated spatial join queries. In this paper, we present SkyQuery, a successful prototype of an evolving federation of astronomy archives. It interoperates using the emerging Web services standard. We describe the SkyQuery architecture and show how it efficiently evaluates a probabilistic federated spatial join query.
cs/0211027
Adaptive Development of Koncepts in Virtual Animats: Insights into the Development of Knowledge
cs.AI
As a part of our effort for studying the evolution and development of cognition, we present results derived from synthetic experimentations in a virtual laboratory where animats develop koncepts adaptively and ground their meaning through action. We introduce the term "koncept" to avoid confusions and ambiguity derived from the wide use of the word "concept". We present the models which our animats use for abstracting koncepts from perceptions, plastically adapt koncepts, and associate koncepts with actions. On a more philosophical vein, we suggest that knowledge is a property of a cognitive system, not an element, and therefore observer-dependent.
cs/0211028
Thinking Adaptive: Towards a Behaviours Virtual Laboratory
cs.AI cs.MA
In this paper we name some of the advantages of virtual laboratories; and propose that a Behaviours Virtual Laboratory should be useful for both biologists and AI researchers, offering a new perspective for understanding adaptive behaviour. We present our development of a Behaviours Virtual Laboratory, which at this stage is focused in action selection, and show some experiments to illustrate the properties of our proposal, which can be accessed via Internet.
cs/0211029
Modelling intracellular signalling networks using behaviour-based systems and the blackboard architecture
cs.MA q-bio.CB
This paper proposes to model the intracellular signalling networks using a fusion of behaviour-based systems and the blackboard architecture. In virtue of this fusion, the model developed by us, which has been named Cellulat, allows to take account two essential aspects of the intracellular signalling networks: (1) the cognitive capabilities of certain types of networks' components and (2) the high level of spatial organization of these networks. A simple example of modelling of Ca2+ signalling pathways using Cellulat is presented here. An intracellular signalling virtual laboratory is being developed from Cellulat.
cs/0211030
Integration of Computational Techniques for the Modelling of Signal Transduction
cs.MA q-bio.CB
A cell can be seen as an adaptive autonomous agent or as a society of adaptive autonomous agents, where each can exhibit a particular behaviour depending on its cognitive capabilities. We present an intracellular signalling model obtained by integrating several computational techniques into an agent-based paradigm. Cellulat, the model, takes into account two essential aspects of the intracellular signalling networks: cognitive capacities and a spatial organization. Exemplifying the functionality of the system by modelling the EGFR signalling pathway, we discuss the methodology as well as the purposes of an intracellular signalling virtual laboratory, presently under development.
cs/0211031
Redundancy in Logic I: CNF Propositional Formulae
cs.AI cs.CC
A knowledge base is redundant if it contains parts that can be inferred from the rest of it. We study the problem of checking whether a CNF formula (a set of clauses) is redundant, that is, it contains clauses that can be derived from the other ones. Any CNF formula can be made irredundant by deleting some of its clauses: what results is an irredundant equivalent subset (I.E.S.) We study the complexity of some related problems: verification, checking existence of a I.E.S. with a given size, checking necessary and possible presence of clauses in I.E.S.'s, and uniqueness. We also consider the problem of redundancy with different definitions of equivalence.
cs/0211033
Propositional satisfiability in declarative programming
cs.LO cs.AI
Answer-set programming (ASP) paradigm is a way of using logic to solve search problems. Given a search problem, to solve it one designs a theory in the logic so that models of this theory represent problem solutions. To compute a solution to a problem one needs to compute a model of the corresponding theory. Several answer-set programming formalisms have been developed on the basis of logic programming with the semantics of stable models. In this paper we show that also the logic of predicate calculus gives rise to effective implementations of the ASP paradigm, similar in spirit to logic programming with stable model semantics and with a similar scope of applicability. Specifically, we propose two logics based on predicate calculus as formalisms for encoding search problems. We show that the expressive power of these logics is given by the class NP-search. We demonstrate how to use them in programming and develop computational tools for model finding. In the case of one of the logics our techniques reduce the problem to that of propositional satisfiability and allow one to use off-the-shelf satisfiability solvers. The language of the other logic has more complex syntax and provides explicit means to model some high-level constraints. For theories in this logic, we designed our own solver that takes advantage of the expanded syntax. We present experimental results demonstrating computational effectiveness of the overall approach.
cs/0211035
Monadic Style Control Constructs for Inference Systems
cs.AI cs.PL
Recent advances in programming languages study and design have established a standard way of grounding computational systems representation in category theory. These formal results led to a better understanding of issues of control and side-effects in functional and imperative languages. Another benefit is a better way of modelling computational effects in logical frameworks. With this analogy in mind, we embark on an investigation of inference systems based on considering inference behaviour as a form of computation. We delineate a categorical formalisation of control constructs in inference systems. This representation emphasises the parallel between the modular articulation of the categorical building blocks (triples) used to account for the inference architecture and the modular composition of cognitive processes.
cs/0211038
Dynamic Adjustment of the Motivation Degree in an Action Selection Mechanism
cs.AI
This paper presents a model for dynamic adjustment of the motivation degree, using a reinforcement learning approach, in an action selection mechanism previously developed by the authors. The learning takes place in the modification of a parameter of the model of combination of internal and external stimuli. Experiments that show the claimed properties are presented, using a VR simulation developed for such purposes. The importance of adaptation by learning in action selection is also discussed.
cs/0211039
Action Selection Properties in a Software Simulated Agent
cs.AI
This article analyses the properties of the Internal Behaviour network, an action selection mechanism previously proposed by the authors, with the aid of a simulation developed for such ends. A brief review of the Internal Behaviour network is followed by the explanation of the implementation of the simulation. Then, experiments are presented and discussed analysing the properties of the action selection in the proposed model.
cs/0211040
A Model for Combination of External and Internal Stimuli in the Action Selection of an Autonomous Agent
cs.AI
This paper proposes a model for combination of external and internal stimuli for the action selection in an autonomous agent, based in an action selection mechanism previously proposed by the authors. This combination model includes additive and multiplicative elements, which allows to incorporate new properties, which enhance the action selection. A given parameter a, which is part of the proposed model, allows to regulate the degree of dependence of the observed external behaviour from the internal states of the entity.
cs/0211041
An Approach to Automatic Indexing of Scientific Publications in High Energy Physics for Database SPIRES HEP
cs.IR cs.DL
We introduce an approach to automatic indexing of e-prints based on a pattern-matching technique making extensive use of an Associative Patterns Dictionary (APD), developed by us. Entries in the APD consist of natural language phrases with the same semantic interpretation as a set of keywords from a controlled vocabulary. The method also allows to recognize within e-prints formulae written in TeX notations that might also appear as keywords. We present an automatic indexing system, AUTEX, which we have applied to keyword index e-prints in selected areas in high energy physics (HEP) making use of the DESY-HEPI thesaurus as a controlled vocabulary.
cs/0211042
Database Repairs and Analytic Tableaux
cs.DB cs.LO
In this article, we characterize in terms of analytic tableaux the repairs of inconsistent relational databases, that is databases that do not satisfy a given set of integrity constraints. For this purpose we provide closing and opening criteria for branches in tableaux that are built for database instances and their integrity constraints. We use the tableaux based characterization as a basis for consistent query answering, that is for retrieving from the database answers to queries that are consistent wrt the integrity constraints.
cs/0212004
Minimal-Change Integrity Maintenance Using Tuple Deletions
cs.DB
We address the problem of minimal-change integrity maintenance in the context of integrity constraints in relational databases. We assume that integrity-restoration actions are limited to tuple deletions. We identify two basic computational issues: repair checking (is a database instance a repair of a given database?) and consistent query answers (is a tuple an answer to a given query in every repair of a given database?). We study the computational complexity of both problems, delineating the boundary between the tractable and the intractable. We consider denial constraints, general functional and inclusion dependencies, as well as key and foreign key constraints. Our results shed light on the computational feasibility of minimal-change integrity maintenance. The tractable cases should lead to practical implementations. The intractability results highlight the inherent limitations of any integrity enforcement mechanism, e.g., triggers or referential constraint actions, as a way of performing minimal-change integrity maintenance.
cs/0212006
Use of openMosix for parallel I/O balancing on storage in Linux cluster
cs.DC cs.DB
In this paper I present some experiences made in the matter of I/O for Linux Clustering. In particular is illustrated the use of the package openMosix, a balancer of workload for processes running in a cluster of nodes. I describe some tests for balancing the load of I/O storage massive processes in a cluster with four components. This work is been written for the proceedings of the workshop Linux cluster: the openMosix approach held at CINECA, Bologna, Italy, on 28 november 2002.
cs/0212008
Principal Manifolds and Nonlinear Dimension Reduction via Local Tangent Space Alignment
cs.LG cs.AI
Nonlinear manifold learning from unorganized data points is a very challenging unsupervised learning and data visualization problem with a great variety of applications. In this paper we present a new algorithm for manifold learning and nonlinear dimension reduction. Based on a set of unorganized data points sampled with noise from the manifold, we represent the local geometry of the manifold using tangent spaces learned by fitting an affine subspace in a neighborhood of each data point. Those tangent spaces are aligned to give the internal global coordinates of the data points with respect to the underlying manifold by way of a partial eigendecomposition of the neighborhood connection matrix. We present a careful error analysis of our algorithm and show that the reconstruction errors are of second-order accuracy. We illustrate our algorithm using curves and surfaces both in 2D/3D and higher dimensional Euclidean spaces, and 64-by-64 pixel face images with various pose and lighting conditions. We also address several theoretical and algorithmic issues for further research and improvements.
cs/0212010
JohnnyVon: Self-Replicating Automata in Continuous Two-Dimensional Space
cs.NE cs.CE
JohnnyVon is an implementation of self-replicating automata in continuous two-dimensional space. Two types of particles drift about in a virtual liquid. The particles are automata with discrete internal states but continuous external relationships. Their internal states are governed by finite state machines but their external relationships are governed by a simulated physics that includes brownian motion, viscosity, and spring-like attractive and repulsive forces. The particles can be assembled into patterns that can encode arbitrary strings of bits. We demonstrate that, if an arbitrary "seed" pattern is put in a "soup" of separate individual particles, the pattern will replicate by assembling the individual particles into copies of itself. We also show that, given sufficient time, a soup of separate individual particles will eventually spontaneously form self-replicating patterns. We discuss the implications of JohnnyVon for research in nanotechnology, theoretical biology, and artificial life.
cs/0212011
Mining the Web for Lexical Knowledge to Improve Keyphrase Extraction: Learning from Labeled and Unlabeled Data
cs.LG cs.IR
Keyphrases are useful for a variety of purposes, including summarizing, indexing, labeling, categorizing, clustering, highlighting, browsing, and searching. The task of automatic keyphrase extraction is to select keyphrases from within the text of a given document. Automatic keyphrase extraction makes it feasible to generate keyphrases for the huge number of documents that do not have manually assigned keyphrases. Good performance on this task has been obtained by approaching it as a supervised learning problem. An input document is treated as a set of candidate phrases that must be classified as either keyphrases or non-keyphrases. To classify a candidate phrase as a keyphrase, the most important features (attributes) appear to be the frequency and location of the candidate phrase in the document. Recent work has demonstrated that it is also useful to know the frequency of the candidate phrase as a manually assigned keyphrase for other documents in the same domain as the given document (e.g., the domain of computer science). Unfortunately, this keyphrase-frequency feature is domain-specific (the learning process must be repeated for each new domain) and training-intensive (good performance requires a relatively large number of training documents in the given domain, with manually assigned keyphrases). The aim of the work described here is to remove these limitations. In this paper, I introduce new features that are derived by mining lexical knowledge from a very large collection of unlabeled data, consisting of approximately 350 million Web pages without manually assigned keyphrases. I present experiments that show that the new features result in improved keyphrase extraction, although they are neither domain-specific nor training-intensive.
cs/0212012
Unsupervised Learning of Semantic Orientation from a Hundred-Billion-Word Corpus
cs.LG cs.IR
The evaluative character of a word is called its semantic orientation. A positive semantic orientation implies desirability (e.g., "honest", "intrepid") and a negative semantic orientation implies undesirability (e.g., "disturbing", "superfluous"). This paper introduces a simple algorithm for unsupervised learning of semantic orientation from extremely large corpora. The method involves issuing queries to a Web search engine and using pointwise mutual information to analyse the results. The algorithm is empirically evaluated using a training corpus of approximately one hundred billion words -- the subset of the Web that is indexed by the chosen search engine. Tested with 3,596 words (1,614 positive and 1,982 negative), the algorithm attains an accuracy of 80%. The 3,596 test words include adjectives, adverbs, nouns, and verbs. The accuracy is comparable with the results achieved by Hatzivassiloglou and McKeown (1997), using a complex four-stage supervised learning algorithm that is restricted to determining the semantic orientation of adjectives.
cs/0212013
Learning to Extract Keyphrases from Text
cs.LG cs.IR
Many academic journals ask their authors to provide a list of about five to fifteen key words, to appear on the first page of each article. Since these key words are often phrases of two or more words, we prefer to call them keyphrases. There is a surprisingly wide variety of tasks for which keyphrases are useful, as we discuss in this paper. Recent commercial software, such as Microsoft's Word 97 and Verity's Search 97, includes algorithms that automatically extract keyphrases from documents. In this paper, we approach the problem of automatically extracting keyphrases from text as a supervised learning task. We treat a document as a set of phrases, which the learning algorithm must learn to classify as positive or negative examples of keyphrases. Our first set of experiments applies the C4.5 decision tree induction algorithm to this learning task. The second set of experiments applies the GenEx algorithm to the task. We developed the GenEx algorithm specifically for this task. The third set of experiments examines the performance of GenEx on the task of metadata generation, relative to the performance of Microsoft's Word 97. The fourth and final set of experiments investigates the performance of GenEx on the task of highlighting, relative to Verity's Search 97. The experimental results support the claim that a specialized learning algorithm (GenEx) can generate better keyphrases than a general-purpose learning algorithm (C4.5) and the non-learning algorithms that are used in commercial software (Word 97 and Search 97).
cs/0212014
Extraction of Keyphrases from Text: Evaluation of Four Algorithms
cs.LG cs.IR
This report presents an empirical evaluation of four algorithms for automatically extracting keywords and keyphrases from documents. The four algorithms are compared using five different collections of documents. For each document, we have a target set of keyphrases, which were generated by hand. The target keyphrases were generated for human readers; they were not tailored for any of the four keyphrase extraction algorithms. Each of the algorithms was evaluated by the degree to which the algorithm's keyphrases matched the manually generated keyphrases. The four algorithms were (1) the AutoSummarize feature in Microsoft's Word 97, (2) an algorithm based on Eric Brill's part-of-speech tagger, (3) the Summarize feature in Verity's Search 97, and (4) NRC's Extractor algorithm. For all five document collections, NRC's Extractor yields the best match with the manually generated keyphrases.
cs/0212015
Answering Subcognitive Turing Test Questions: A Reply to French
cs.CL
Robert French has argued that a disembodied computer is incapable of passing a Turing Test that includes subcognitive questions. Subcognitive questions are designed to probe the network of cultural and perceptual associations that humans naturally develop as we live, embodied and embedded in the world. In this paper, I show how it is possible for a disembodied computer to answer subcognitive questions appropriately, contrary to French's claim. My approach to answering subcognitive questions is to use statistical information extracted from a very large collection of text. In particular, I show how it is possible to answer a sample of subcognitive questions taken from French, by issuing queries to a search engine that indexes about 350 million Web pages. This simple algorithm may shed light on the nature of human (sub-) cognition, but the scope of this paper is limited to demonstrating that French is mistaken: a disembodied computer can answer subcognitive questions.
cs/0212017
Classes of Spatiotemporal Objects and Their Closure Properties
cs.DB
We present a data model for spatio-temporal databases. In this model spatio-temporal data is represented as a finite union of objects described by means of a spatial reference object, a temporal object and a geometric transformation function that determines the change or movement of the reference object in time. We define a number of practically relevant classes of spatio-temporal objects, and give complete results concerning closure under Boolean set operators for these classes. Since only few classes are closed under all set operators, we suggest an extension of the model, which leads to better closure properties, and therefore increased practical applicability. We also discuss a normal form for this extended data model.
cs/0212018
Real numbers having ultimately periodic representations in abstract numeration systems
cs.CC cs.CL
Using a genealogically ordered infinite regular language, we know how to represent an interval of R. Numbers having an ultimately periodic representation play a special role in classical numeration systems. The aim of this paper is to characterize the numbers having an ultimately periodic representation in generalized systems built on a regular language. The syntactical properties of these words are also investigated. Finally, we show the equivalence of the classical "theta"-expansions with our generalized representations in some special case related to a Pisot number "theta".
cs/0212019
Thinking, Learning, and Autonomous Problem Solving
cs.NE
Ever increasing computational power will require methods for automatic programming. We present an alternative to genetic programming, based on a general model of thinking and learning. The advantage is that evolution takes place in the space of constructs and can thus exploit the mathematical structures of this space. The model is formalized, and a macro language is presented which allows for a formal yet intuitive description of the problem under consideration. A prototype has been developed to implement the scheme in PERL. This method will lead to a concentration on the analysis of problems, to a more rapid prototyping, to the treatment of new problem classes, and to the investigation of philosophical problems. We see fields of application in nonlinear differential equations, pattern recognition, robotics, model building, and animated pictures.
cs/0212020
Learning Algorithms for Keyphrase Extraction
cs.LG cs.CL cs.IR
Many academic journals ask their authors to provide a list of about five to fifteen keywords, to appear on the first page of each article. Since these key words are often phrases of two or more words, we prefer to call them keyphrases. There is a wide variety of tasks for which keyphrases are useful, as we discuss in this paper. We approach the problem of automatically extracting keyphrases from text as a supervised learning task. We treat a document as a set of phrases, which the learning algorithm must learn to classify as positive or negative examples of keyphrases. Our first set of experiments applies the C4.5 decision tree induction algorithm to this learning task. We evaluate the performance of nine different configurations of C4.5. The second set of experiments applies the GenEx algorithm to the task. We developed the GenEx algorithm specifically for automatically extracting keyphrases from text. The experimental results support the claim that a custom-designed algorithm (GenEx), incorporating specialized procedural domain knowledge, can generate better keyphrases than a generalpurpose algorithm (C4.5). Subjective human evaluation of the keyphrases generated by Extractor suggests that about 80% of the keyphrases are acceptable to human readers. This level of performance should be satisfactory for a wide variety of applications.
cs/0212021
A Simple Model of Unbounded Evolutionary Versatility as a Largest-Scale Trend in Organismal Evolution
cs.NE cs.CE q-bio.PE
The idea that there are any large-scale trends in the evolution of biological organisms is highly controversial. It is commonly believed, for example, that there is a large-scale trend in evolution towards increasing complexity, but empirical and theoretical arguments undermine this belief. Natural selection results in organisms that are well adapted to their local environments, but it is not clear how local adaptation can produce a global trend. In this paper, I present a simple computational model, in which local adaptation to a randomly changing environment results in a global trend towards increasing evolutionary versatility. In this model, for evolutionary versatility to increase without bound, the environment must be highly dynamic. The model also shows that unbounded evolutionary versatility implies an accelerating evolutionary pace. I believe that unbounded increase in evolutionary versatility is a large-scale trend in evolution. I discuss some of the testable predictions about organismal evolution that are suggested by the model.
cs/0212022
Algorithms for Rapidly Dispersing Robot Swarms in Unknown Environments
cs.RO
We develop and analyze algorithms for dispersing a swarm of primitive robots in an unknown environment, R. The primary objective is to minimize the makespan, that is, the time to fill the entire region. An environment is composed of pixels that form a connected subset of the integer grid. There is at most one robot per pixel and robots move horizontally or vertically at unit speed. Robots enter R by means of k>=1 door pixels Robots are primitive finite automata, only having local communication, local sensors, and a constant-sized memory. We first give algorithms for the single-door case (i.e., k=1), analyzing the algorithms both theoretically and experimentally. We prove that our algorithms have optimal makespan 2A-1, where A is the area of R. We next give an algorithm for the multi-door case (k>1), based on a wall-following version of the leader-follower strategy. We prove that our strategy is O(log(k+1))-competitive, and that this bound is tight for our strategy and other related strategies.
cs/0212023
How to Shift Bias: Lessons from the Baldwin Effect
cs.LG cs.NE
An inductive learning algorithm takes a set of data as input and generates a hypothesis as output. A set of data is typically consistent with an infinite number of hypotheses; therefore, there must be factors other than the data that determine the output of the learning algorithm. In machine learning, these other factors are called the bias of the learner. Classical learning algorithms have a fixed bias, implicit in their design. Recently developed learning algorithms dynamically adjust their bias as they search for a hypothesis. Algorithms that shift bias in this manner are not as well understood as classical algorithms. In this paper, we show that the Baldwin effect has implications for the design and analysis of bias shifting algorithms. The Baldwin effect was proposed in 1896, to explain how phenomena that might appear to require Lamarckian evolution (inheritance of acquired characteristics) can arise from purely Darwinian evolution. Hinton and Nowlan presented a computational model of the Baldwin effect in 1987. We explore a variation on their model, which we constructed explicitly to illustrate the lessons that the Baldwin effect has for research in bias shifting algorithms. The main lesson is that it appears that a good strategy for shift of bias in a learning algorithm is to begin with a weak bias and gradually shift to a strong bias.
cs/0212024
Unsupervised Language Acquisition: Theory and Practice
cs.CL cs.LG
In this thesis I present various algorithms for the unsupervised machine learning of aspects of natural languages using a variety of statistical models. The scientific object of the work is to examine the validity of the so-called Argument from the Poverty of the Stimulus advanced in favour of the proposition that humans have language-specific innate knowledge. I start by examining an a priori argument based on Gold's theorem, that purports to prove that natural languages cannot be learned, and some formal issues related to the choice of statistical grammars rather than symbolic grammars. I present three novel algorithms for learning various parts of natural languages: first, an algorithm for the induction of syntactic categories from unlabelled text using distributional information, that can deal with ambiguous and rare words; secondly, a set of algorithms for learning morphological processes in a variety of languages, including languages such as Arabic with non-concatenative morphology; thirdly an algorithm for the unsupervised induction of a context-free grammar from tagged text. I carefully examine the interaction between the various components, and show how these algorithms can form the basis for a empiricist model of language acquisition. I therefore conclude that the Argument from the Poverty of the Stimulus is unsupported by the evidence.
cs/0212025
Searching for Plannable Domains can Speed up Reinforcement Learning
cs.AI
Reinforcement learning (RL) involves sequential decision making in uncertain environments. The aim of the decision-making agent is to maximize the benefit of acting in its environment over an extended period of time. Finding an optimal policy in RL may be very slow. To speed up learning, one often used solution is the integration of planning, for example, Sutton's Dyna algorithm, or various other methods using macro-actions. Here we suggest to separate plannable, i.e., close to deterministic parts of the world, and focus planning efforts in this domain. A novel reinforcement learning method called plannable RL (pRL) is proposed here. pRL builds a simple model, which is used to search for macro actions. The simplicity of the model makes planning computationally inexpensive. It is shown that pRL finds an optimal policy, and that plannable macro actions found by pRL are near-optimal. In turn, it is unnecessary to try large numbers of macro actions, which enables fast learning. The utility of pRL is demonstrated by computer simulations.
cs/0212027
Qualitative Study of a Robot Arm as a Hamiltonian System
cs.RO
A double pendulum subject to external torques is used as a model to study the stability of a planar manipulator with two links and two rotational driven joints. The hamiltonian equations of motion and the fixed points (stationary solutions) in phase space are determined. Under suitable conditions, the presence of constant torques does not change the number of fixed points, and preserves the topology of orbits in their linear neighborhoods; two equivalent invariant manifolds are observed, each corresponding to a saddle-center fixed point.
cs/0212028
Technical Note: Bias and the Quantification of Stability
cs.LG cs.CV
Research on bias in machine learning algorithms has generally been concerned with the impact of bias on predictive accuracy. We believe that there are other factors that should also play a role in the evaluation of bias. One such factor is the stability of the algorithm; in other words, the repeatability of the results. If we obtain two sets of data from the same phenomenon, with the same underlying probability distribution, then we would like our learning algorithm to induce approximately the same concepts from both sets of data. This paper introduces a method for quantifying stability, based on a measure of the agreement between concepts. We also discuss the relationships among stability, predictive accuracy, and bias.
cs/0212029
A Theory of Cross-Validation Error
cs.LG cs.CV
This paper presents a theory of error in cross-validation testing of algorithms for predicting real-valued attributes. The theory justifies the claim that predicting real-valued attributes requires balancing the conflicting demands of simplicity and accuracy. Furthermore, the theory indicates precisely how these conflicting demands must be balanced, in order to minimize cross-validation error. A general theory is presented, then it is developed in detail for linear regression and instance-based learning.
cs/0212030
Theoretical Analyses of Cross-Validation Error and Voting in Instance-Based Learning
cs.LG cs.CV
This paper begins with a general theory of error in cross-validation testing of algorithms for supervised learning from examples. It is assumed that the examples are described by attribute-value pairs, where the values are symbolic. Cross-validation requires a set of training examples and a set of testing examples. The value of the attribute that is to be predicted is known to the learner in the training set, but unknown in the testing set. The theory demonstrates that cross-validation error has two components: error on the training set (inaccuracy) and sensitivity to noise (instability). This general theory is then applied to voting in instance-based learning. Given an example in the testing set, a typical instance-based learning algorithm predicts the designated attribute by voting among the k nearest neighbors (the k most similar examples) to the testing example in the training set. Voting is intended to increase the stability (resistance to noise) of instance-based learning, but a theoretical analysis shows that there are circumstances in which voting can be destabilizing. The theory suggests ways to minimize cross-validation error, by insuring that voting is stable and does not adversely affect accuracy.
cs/0212031
Contextual Normalization Applied to Aircraft Gas Turbine Engine Diagnosis
cs.LG cs.CE cs.CV
Diagnosing faults in aircraft gas turbine engines is a complex problem. It involves several tasks, including rapid and accurate interpretation of patterns in engine sensor data. We have investigated contextual normalization for the development of a software tool to help engine repair technicians with interpretation of sensor data. Contextual normalization is a new strategy for employing machine learning. It handles variation in data that is due to contextual factors, rather than the health of the engine. It does this by normalizing the data in a context-sensitive manner. This learning strategy was developed and tested using 242 observations of an aircraft gas turbine engine in a test cell, where each observation consists of roughly 12,000 numbers, gathered over a 12 second interval. There were eight classes of observations: seven deliberately implanted classes of faults and a healthy class. We compared two approaches to implementing our learning strategy: linear regression and instance-based learning. We have three main results. (1) For the given problem, instance-based learning works better than linear regression. (2) For this problem, contextual normalization works better than other common forms of normalization. (3) The algorithms described here can be the basis for a useful software tool for assisting technicians with the interpretation of sensor data.
cs/0212032
Thumbs Up or Thumbs Down? Semantic Orientation Applied to Unsupervised Classification of Reviews
cs.LG cs.CL cs.IR
This paper presents a simple unsupervised learning algorithm for classifying reviews as recommended (thumbs up) or not recommended (thumbs down). The classification of a review is predicted by the average semantic orientation of the phrases in the review that contain adjectives or adverbs. A phrase has a positive semantic orientation when it has good associations (e.g., "subtle nuances") and a negative semantic orientation when it has bad associations (e.g., "very cavalier"). In this paper, the semantic orientation of a phrase is calculated as the mutual information between the given phrase and the word "excellent" minus the mutual information between the given phrase and the word "poor". A review is classified as recommended if the average semantic orientation of its phrases is positive. The algorithm achieves an average accuracy of 74% when evaluated on 410 reviews from Epinions, sampled from four different domains (reviews of automobiles, banks, movies, and travel destinations). The accuracy ranges from 84% for automobile reviews to 66% for movie reviews.
cs/0212033
Mining the Web for Synonyms: PMI-IR versus LSA on TOEFL
cs.LG cs.CL cs.IR
This paper presents a simple unsupervised learning algorithm for recognizing synonyms, based on statistical data acquired by querying a Web search engine. The algorithm, called PMI-IR, uses Pointwise Mutual Information (PMI) and Information Retrieval (IR) to measure the similarity of pairs of words. PMI-IR is empirically evaluated using 80 synonym test questions from the Test of English as a Foreign Language (TOEFL) and 50 synonym test questions from a collection of tests for students of English as a Second Language (ESL). On both tests, the algorithm obtains a score of 74%. PMI-IR is contrasted with Latent Semantic Analysis (LSA), which achieves a score of 64% on the same 80 TOEFL questions. The paper discusses potential applications of the new unsupervised learning algorithm and some implications of the results for LSA and LSI (Latent Semantic Indexing).
cs/0212034
Types of Cost in Inductive Concept Learning
cs.LG cs.CV
Inductive concept learning is the task of learning to assign cases to a discrete set of classes. In real-world applications of concept learning, there are many different types of cost involved. The majority of the machine learning literature ignores all types of cost (unless accuracy is interpreted as a type of cost measure). A few papers have investigated the cost of misclassification errors. Very few papers have examined the many other types of cost. In this paper, we attempt to create a taxonomy of the different types of cost that are involved in inductive concept learning. This taxonomy may help to organize the literature on cost-sensitive learning. We hope that it will inspire researchers to investigate all types of cost in inductive concept learning in more depth.
cs/0212035
Exploiting Context When Learning to Classify
cs.LG cs.CV
This paper addresses the problem of classifying observations when features are context-sensitive, specifically when the testing set involves a context that is different from the training set. The paper begins with a precise definition of the problem, then general strategies are presented for enhancing the performance of classification algorithms on this type of problem. These strategies are tested on two domains. The first domain is the diagnosis of gas turbine engines. The problem is to diagnose a faulty engine in one context, such as warm weather, when the fault has previously been seen only in another context, such as cold weather. The second domain is speech recognition. The problem is to recognize words spoken by a new speaker, not represented in the training set. For both domains, exploiting context results in substantially more accurate classification.
cs/0212036
Myths and Legends of the Baldwin Effect
cs.LG cs.NE
This position paper argues that the Baldwin effect is widely misunderstood by the evolutionary computation community. The misunderstandings appear to fall into two general categories. Firstly, it is commonly believed that the Baldwin effect is concerned with the synergy that results when there is an evolving population of learning individuals. This is only half of the story. The full story is more complicated and more interesting. The Baldwin effect is concerned with the costs and benefits of lifetime learning by individuals in an evolving population. Several researchers have focussed exclusively on the benefits, but there is much to be gained from attention to the costs. This paper explains the two sides of the story and enumerates ten of the costs and benefits of lifetime learning by individuals in an evolving population. Secondly, there is a cluster of misunderstandings about the relationship between the Baldwin effect and Lamarckian inheritance of acquired characteristics. The Baldwin effect is not Lamarckian. A Lamarckian algorithm is not better for most evolutionary computing problems than a Baldwinian algorithm. Finally, Lamarckian inheritance is not a better model of memetic (cultural) evolution than the Baldwin effect.
cs/0212037
The Management of Context-Sensitive Features: A Review of Strategies
cs.LG cs.CV
In this paper, we review five heuristic strategies for handling context-sensitive features in supervised machine learning from examples. We discuss two methods for recovering lost (implicit) contextual information. We mention some evidence that hybrid strategies can have a synergetic effect. We then show how the work of several machine learning researchers fits into this framework. While we do not claim that these strategies exhaust the possibilities, it appears that the framework includes all of the techniques that can be found in the published literature on contextsensitive learning.
cs/0212038
The Identification of Context-Sensitive Features: A Formal Definition of Context for Concept Learning
cs.LG cs.CV
A large body of research in machine learning is concerned with supervised learning from examples. The examples are typically represented as vectors in a multi-dimensional feature space (also known as attribute-value descriptions). A teacher partitions a set of training examples into a finite number of classes. The task of the learning algorithm is to induce a concept from the training examples. In this paper, we formally distinguish three types of features: primary, contextual, and irrelevant features. We also formally define what it means for one feature to be context-sensitive to another feature. Context-sensitive features complicate the task of the learner and potentially impair the learner's performance. Our formal definitions make it possible for a learner to automatically identify context-sensitive features. After context-sensitive features have been identified, there are several strategies that the learner can employ for managing the features; however, a discussion of these strategies is outside of the scope of this paper. The formal definitions presented here correct a flaw in previously proposed definitions. We discuss the relationship between our work and a formal definition of relevance.
cs/0212039
Low Size-Complexity Inductive Logic Programming: The East-West Challenge Considered as a Problem in Cost-Sensitive Classification
cs.LG cs.NE
The Inductive Logic Programming community has considered proof-complexity and model-complexity, but, until recently, size-complexity has received little attention. Recently a challenge was issued "to the international computing community" to discover low size-complexity Prolog programs for classifying trains. The challenge was based on a problem first proposed by Ryszard Michalski, 20 years ago. We interpreted the challenge as a problem in cost-sensitive classification and we applied a recently developed cost-sensitive classifier to the competition. Our algorithm was relatively successful (we won a prize). This paper presents our algorithm and analyzes the results of the competition.
cs/0212040
Data Engineering for the Analysis of Semiconductor Manufacturing Data
cs.LG cs.CE cs.CV
We have analyzed manufacturing data from several different semiconductor manufacturing plants, using decision tree induction software called Q-YIELD. The software generates rules for predicting when a given product should be rejected. The rules are intended to help the process engineers improve the yield of the product, by helping them to discover the causes of rejection. Experience with Q-YIELD has taught us the importance of data engineering -- preprocessing the data to enable or facilitate decision tree induction. This paper discusses some of the data engineering problems we have encountered with semiconductor manufacturing data. The paper deals with two broad classes of problems: engineering the features in a feature vector representation and engineering the definition of the target concept (the classes). Manufacturing process data present special problems for feature engineering, since the data have multiple levels of granularity (detail, resolution). Engineering the target concept is important, due to our focus on understanding the past, as opposed to the more common focus in machine learning on predicting the future.
cs/0212041
Robust Classification with Context-Sensitive Features
cs.LG cs.CV
This paper addresses the problem of classifying observations when features are context-sensitive, especially when the testing set involves a context that is different from the training set. The paper begins with a precise definition of the problem, then general strategies are presented for enhancing the performance of classification algorithms on this type of problem. These strategies are tested on three domains. The first domain is the diagnosis of gas turbine engines. The problem is to diagnose a faulty engine in one context, such as warm weather, when the fault has previously been seen only in another context, such as cold weather. The second domain is speech recognition. The context is given by the identity of the speaker. The problem is to recognize words spoken by a new speaker, not represented in the training set. The third domain is medical prognosis. The problem is to predict whether a patient with hepatitis will live or die. The context is the age of the patient. For all three domains, exploiting context results in substantially more accurate classification.
cs/0212042
Increasing Evolvability Considered as a Large-Scale Trend in Evolution
cs.NE cs.CE q-bio.PE
Evolvability is the capacity to evolve. This paper introduces a simple computational model of evolvability and demonstrates that, under certain conditions, evolvability can increase indefinitely, even when there is no direct selection for evolvability. The model shows that increasing evolvability implies an accelerating evolutionary pace. It is suggested that the conditions for indefinitely increasing evolvability are satisfied in biological and cultural evolution. We claim that increasing evolvability is a large-scale trend in evolution. This hypothesis leads to testable predictions about biological and cultural evolution.
cs/0212045
Local Community Identification through User Access Patterns
cs.IR cs.HC
Community identification algorithms have been used to enhance the quality of the services perceived by its users. Although algorithms for community have a widespread use in the Web, their application to portals or specific subsets of the Web has not been much studied. In this paper, we propose a technique for local community identification that takes into account user access behavior derived from access logs of servers in the Web. The technique takes a departure from the existing community algorithms since it changes the focus of in terest, moving from authors to users. Our approach does not use relations imposed by authors (e.g. hyperlinks in the case of Web pages). It uses information derived from user accesses to a service in order to infer relationships. The communities identified are of great interest to content providers since they can be used to improve quality of their services. We also propose an evaluation methodology for analyzing the results obtained by the algorithm. We present two case studies based on actual data from two services: an online bookstore and an online radio. The case of the online radio is particularly relevant, because it emphasizes the contribution of the proposed algorithm to find out communities in an environment (i.e., streaming media service) without links, that represent the relations imposed by authors (e.g. hyperlinks in the case of Web pages).
cs/0212049
An Ehrenfeucht-Fraisse Game Approach to Collapse Results in Database Theory
cs.LO cs.DB
We present a new Ehrenfeucht-Fraisse game approach to collapse results in database theory and we show that, in principle, this approach suffices to prove every natural generic collapse result. Following this approach we can deal with certain infinite databases where previous, highly involved methods fail. We prove the natural generic collapse for Z-embeddable databases over any linearly ordered context structure with arbitrary monadic predicates, and for N-embeddable databases over the context structure (R,<,+,Mon_Q,Groups). Here, N, Z, R, denote the sets of natural numbers, integers, and real numbers, respectively. Groups is the collection of all subgroups of (R,+) that contain Z, and Mon_Q is the collection of all subsets of a particular infinite subset Q of N. Restricting the complexity of the formulas that may be used to formulate queries to Boolean combinations of purely existential first-order formulas, we even obtain the collapse for N-embeddable databases over any linearly ordered context structure with arbitrary predicates. Finally, we develop the notion of N-representable databases, which is a natural generalization of the classical notion of finitely representable databases. We show that natural generic collapse results for N-embeddable databases can be lifted to the larger class of N-representable databases. To obtain, in particular, the collapse result for (N,<,+,Mon_Q), we explicitly construct a winning strategy for the duplicator in the presence of the built-in addition relation +. This, as a side product, also leads to an Ehrenfeucht-Fraisse game proof of the theorem of Ginsburg and Spanier, stating that the spectra of FO(<,+)-sentences are semi-linear.
cs/0212051
ExploitingWeb Service Semantics: Taxonomies vs. Ontologies
cs.DB
Comprehensive semantic descriptions of Web services are essential to exploit them in their full potential, that is, discovering them dynamically, and enabling automated service negotiation, composition and monitoring. The semantic mechanisms currently available in service registries which are based on taxonomies fail to provide the means to achieve this. Although the terms taxonomy and ontology are sometimes used interchangably there is a critical difference. A taxonomy indicates only class/subclass relationship whereas an ontology describes a domain completely. The essential mechanisms that ontology languages provide include their formal specification (which allows them to be queried) and their ability to define properties of classes. Through properties very accurate descriptions of services can be defined and services can be related to other services or resources. In this paper, we discuss the advantages of describing service semantics through ontology languages and describe how to relate the semantics defined with the services advertised in service registries like UDDI and ebXML.
cs/0212052
Improving the Functionality of UDDI Registries through Web Service Semantics
cs.DB
In this paper we describe a framework for exploiting the semantics of Web services through UDDI registries. As a part of this framework, we extend the DAML-S upper ontology to describe the functionality we find essential for e-businesses. This functionality includes relating the services with electronic catalogs, describing the complementary services and finding services according to the properties of products or services. Once the semantics is defined, there is a need for a mechanism in the service registry to relate it with the service advertised. The ontology model developed is general enough to be used with any service registry. However when it comes to relating the semantics with services advertised, the capabilities provided by the registry effects how this is achieved. We demonstrate how to integrate the described service semantics to UDDI registries.
cs/0212053
Merging Locally Correct Knowledge Bases: A Preliminary Report
cs.AI cs.LO
Belief integration methods are often aimed at deriving a single and consistent knowledge base that retains as much as possible of the knowledge bases to integrate. The rationale behind this approach is the minimal change principle: the result of the integration process should differ as less as possible from the knowledge bases to integrate. We show that this principle can be reformulated in terms of a more general model of belief revision, based on the assumption that inconsistency is due to the mistakes the knowledge bases contain. Current belief revision strategies are based on a specific kind of mistakes, which however does not include all possible ones. Some alternative possibilities are discussed.
cs/0301001
Least squares fitting of circles and lines
cs.CV
We study theoretical and computational aspects of the least squares fit (LSF) of circles and circular arcs. First we discuss the existence and uniqueness of LSF and various parametrization schemes. Then we evaluate several popular circle fitting algorithms and propose a new one that surpasses the existing methods in reliability. We also discuss and compare direct (algebraic) circle fits.
cs/0301006
Temporal plannability by variance of the episode length
cs.AI
Optimization of decision problems in stochastic environments is usually concerned with maximizing the probability of achieving the goal and minimizing the expected episode length. For interacting agents in time-critical applications, learning of the possibility of scheduling of subtasks (events) or the full task is an additional relevant issue. Besides, there exist highly stochastic problems where the actual trajectories show great variety from episode to episode, but completing the task takes almost the same amount of time. The identification of sub-problems of this nature may promote e.g., planning, scheduling and segmenting Markov decision processes. In this work, formulae for the average duration as well as the standard deviation of the duration of events are derived. The emerging Bellman-type equation is a simple extension of Sobel's work (1982). Methods of dynamic programming as well as methods of reinforcement learning can be applied for our extension. Computer demonstration on a toy problem serve to highlight the principle.
cs/0301007
Kalman filter control in the reinforcement learning framework
cs.LG cs.AI
There is a growing interest in using Kalman-filter models in brain modelling. In turn, it is of considerable importance to make Kalman-filters amenable for reinforcement learning. In the usual formulation of optimal control it is computed off-line by solving a backward recursion. In this technical note we show that slight modification of the linear-quadratic-Gaussian Kalman-filter model allows the on-line estimation of optimal control and makes the bridge to reinforcement learning. Moreover, the learning rule for value estimation assumes a Hebbian form weighted by the error of the value estimation.
cs/0301008
Formal Concept Analysis and Resolution in Algebraic Domains
cs.LO cs.AI
We relate two formerly independent areas: Formal concept analysis and logic of domains. We will establish a correspondene between contextual attribute logic on formal contexts resp. concept lattices and a clausal logic on coherent algebraic cpos. We show how to identify the notion of formal concept in the domain theoretic setting. In particular, we show that a special instance of the resolution rule from the domain logic coincides with the concept closure operator from formal concept analysis. The results shed light on the use of contexts and domains for knowledge representation and reasoning purposes.
cs/0301009
A Script Language for Data Integration in Database
cs.DB
A Script Language in this paper is designed to transform the original data into the target data by the computing formula. The Script Language can be translated into the corresponding SQL Language, and the computation is finally implemented by the first type of dynamic SQL. The Script Language has the operations of insert, update, delete, union, intersect, and minus for the table in the database.The Script Language is edited by a text file and you can easily modify the computing formula in the text file to deal with the situations when the computing formula have been changed. So you only need modify the text of the script language, but needn't change the programs that have complied.
cs/0301010
Comparisons and Computation of Well-founded Semantics for Disjunctive Logic Programs
cs.AI
Much work has been done on extending the well-founded semantics to general disjunctive logic programs and various approaches have been proposed. However, these semantics are different from each other and no consensus is reached about which semantics is the most intended. In this paper we look at disjunctive well-founded reasoning from different angles. We show that there is an intuitive form of the well-founded reasoning in disjunctive logic programming which can be characterized by slightly modifying some exisitng approaches to defining disjunctive well-founded semantics, including program transformations, argumentation, unfounded sets (and resolution-like procedure). We also provide a bottom-up procedure for this semantics. The significance of our work is not only in clarifying the relationship among different approaches, but also shed some light on what is an intended well-founded semantics for disjunctive logic programs.
cs/0301014
Convergence and Loss Bounds for Bayesian Sequence Prediction
cs.LG cs.AI math.PR
The probability of observing $x_t$ at time $t$, given past observations $x_1...x_{t-1}$ can be computed with Bayes' rule if the true generating distribution $\mu$ of the sequences $x_1x_2x_3...$ is known. If $\mu$ is unknown, but known to belong to a class $M$ one can base ones prediction on the Bayes mix $\xi$ defined as a weighted sum of distributions $\nu\in M$. Various convergence results of the mixture posterior $\xi_t$ to the true posterior $\mu_t$ are presented. In particular a new (elementary) derivation of the convergence $\xi_t/\mu_t\to 1$ is provided, which additionally gives the rate of convergence. A general sequence predictor is allowed to choose an action $y_t$ based on $x_1...x_{t-1}$ and receives loss $\ell_{x_t y_t}$ if $x_t$ is the next symbol of the sequence. No assumptions are made on the structure of $\ell$ (apart from being bounded) and $M$. The Bayes-optimal prediction scheme $\Lambda_\xi$ based on mixture $\xi$ and the Bayes-optimal informed prediction scheme $\Lambda_\mu$ are defined and the total loss $L_\xi$ of $\Lambda_\xi$ is bounded in terms of the total loss $L_\mu$ of $\Lambda_\mu$. It is shown that $L_\xi$ is bounded for bounded $L_\mu$ and $L_\xi/L_\mu\to 1$ for $L_\mu\to \infty$. Convergence of the instantaneous losses are also proven.
cs/0301017
Completeness and Decidability Properties for Functional Dependencies in XML
cs.DB
XML is of great importance in information storage and retrieval because of its recent emergence as a standard for data representation and interchange on the Internet. However XML provides little semantic content and as a result several papers have addressed the topic of how to improve the semantic expressiveness of XML. Among the most important of these approaches has been that of defining integrity constraints in XML. In a companion paper we defined strong functional dependencies in XML(XFDs). We also presented a set of axioms for reasoning about the implication of XFDs and showed that the axiom system is sound for arbitrary XFDs. In this paper we prove that the axioms are also complete for unary XFDs (XFDs with a single path on the l.h.s.). The second contribution of the paper is to prove that the implication problem for unary XFDs is decidable and to provide a linear time algorithm for it.
cs/0301018
Novel Runtime Systems Support for Adaptive Compositional Modeling on the Grid
cs.CE cs.DC
Grid infrastructures and computing environments have progressed significantly in the past few years. The vision of truly seamless Grid usage relies on runtime systems support that is cognizant of the operational issues underlying grid computations and, at the same time, is flexible enough to accommodate diverse application scenarios. This paper addresses the twin aspects of Grid infrastructure and application support through a novel combination of two computational technologies: Weaves - a source-language independent parallel runtime compositional framework that operates through reverse-analysis of compiled object files, and runtime recommender systems that aid in dynamic knowledge-based application composition. Domain-specific adaptivity is exploited through a novel compositional system that supports runtime recommendation of code modules and a sophisticated checkpointing and runtime migration solution that can be transparently deployed over Grid infrastructures. A core set of "adaptivity schemas" are provided as templates for adaptive composition of large-scale scientific computations. Implementation issues, motivating application contexts, and preliminary results are described.
cs/0301023
A semantic framework for preference handling in answer set programming
cs.AI
We provide a semantic framework for preference handling in answer set programming. To this end, we introduce preference preserving consequence operators. The resulting fixpoint characterizations provide us with a uniform semantic framework for characterizing preference handling in existing approaches. Although our approach is extensible to other semantics by means of an alternating fixpoint theory, we focus here on the elaboration of preferences under answer set semantics. Alternatively, we show how these approaches can be characterized by the concept of order preservation. These uniform semantic characterizations provide us with new insights about interrelationships and moreover about ways of implementation.
cs/0302001
Many Hard Examples in Exact Phase Transitions with Application to Generating Hard Satisfiable Instances
cs.CC cond-mat.stat-mech cs.AI cs.DM
This paper first analyzes the resolution complexity of two random CSP models (i.e. Model RB/RD) for which we can establish the existence of phase transitions and identify the threshold points exactly. By encoding CSPs into CNF formulas, it is proved that almost all instances of Model RB/RD have no tree-like resolution proofs of less than exponential size. Thus, we not only introduce new families of CNF formulas hard for resolution, which is a central task of Proof-Complexity theory, but also propose models with both many hard instances and exact phase transitions. Then, the implications of such models are addressed. It is shown both theoretically and experimentally that an application of Model RB/RD might be in the generation of hard satisfiable instances, which is not only of practical importance but also related to some open problems in cryptography such as generating one-way functions. Subsequently, a further theoretical support for the generation method is shown by establishing exponential lower bounds on the complexity of solving random satisfiable and forced satisfiable instances of RB/RD near the threshold. Finally, conclusions are presented, as well as a detailed comparison of Model RB/RD with the Hamiltonian cycle problem and random 3-SAT, which, respectively, exhibit three different kinds of phase transition behavior in NP-complete problems.
cs/0302002
Optimizing GoTools' Search Heuristics using Genetic Algorithms
cs.NE
GoTools is a program which solves life & death problems in the game of Go. This paper describes experiments using a Genetic Algorithm to optimize heuristic weights used by GoTools' tree-search. The complete set of heuristic weights is composed of different subgroups, each of which can be optimized with a suitable fitness function. As a useful side product, an MPI interface for FreePascal was implemented to allow the use of a parallelized fitness function running on a Beowulf cluster. The aim of this exercise is to optimize the current version of GoTools, and to make tools available in preparation of an extension of GoTools for solving open boundary life & death problems, which will introduce more heuristic parameters to be fine tuned.
cs/0302004
Unique Pattern Matching in Strings
cs.PL cs.DB
Regular expression patterns are a key feature of document processing languages like Perl and XDuce. It is in this context that the first and longest match policies have been proposed to disambiguate the pattern matching process. We formally define a matching semantics with these policies and show that the generally accepted method of simulating longest match by first match and recursion is incorrect. We continue by solving the associated type inference problem, which consists in calculating for every subexpression the set of words the subexpression can still match when these policies are in effect, and show how this algorithm can be used to efficiently implement the matching process.
cs/0302012
The New AI: General & Sound & Relevant for Physics
cs.AI cs.LG quant-ph
Most traditional artificial intelligence (AI) systems of the past 50 years are either very limited, or based on heuristics, or both. The new millennium, however, has brought substantial progress in the field of theoretically optimal and practically feasible algorithms for prediction, search, inductive inference based on Occam's razor, problem solving, decision making, and reinforcement learning in environments of a very general type. Since inductive inference is at the heart of all inductive sciences, some of the results are relevant not only for AI and computer science but also for physics, provoking nontraditional predictions based on Zuse's thesis of the computer-generated universe.
cs/0302014
An Algorithm for Aligning Sentences in Bilingual Corpora Using Lexical Information
cs.CL
In this paper we describe an algorithm for aligning sentences with their translations in a bilingual corpus using lexical information of the languages. Existing efficient algorithms ignore word identities and consider only the sentence lengths (Brown, 1991; Gale and Church, 1993). For a sentence in the source language text, the proposed algorithm picks the most likely translation from the target language text using lexical information and certain heuristics. It does not do statistical analysis using sentence lengths. The algorithm is language independent. It also aids in detecting addition and deletion of text in translations. The algorithm gives comparable results with the existing algorithms in most of the cases while it does better in cases where statistical algorithms do not give good results.
cs/0302015
Unsupervised Learning in a Framework of Information Compression by Multiple Alignment, Unification and Search
cs.AI cs.LG
This paper describes a novel approach to unsupervised learning that has been developed within a framework of "information compression by multiple alignment, unification and search" (ICMAUS), designed to integrate learning with other AI functions such as parsing and production of language, fuzzy pattern recognition, probabilistic and exact forms of reasoning, and others.
cs/0302021
Building an Open Language Archives Community on the OAI Foundation
cs.CL cs.DL
The Open Language Archives Community (OLAC) is an international partnership of institutions and individuals who are creating a worldwide virtual library of language resources. The Dublin Core (DC) Element Set and the OAI Protocol have provided a solid foundation for the OLAC framework. However, we need more precision in community-specific aspects of resource description than is offered by DC. Furthermore, many of the institutions and individuals who might participate in OLAC do not have the technical resources to support the OAI protocol. This paper presents our solutions to these two problems. To address the first, we have developed an extensible application profile for language resource metadata. To address the second, we have implemented Vida (the virtual data provider) and Viser (the virtual service provider), which permit community members to provide data and services without having to implement the OAI protocol. These solutions are generic and could be adopted by other specialized subcommunities.
cs/0302023
Segmentation, Indexing, and Visualization of Extended Instructional Videos
cs.IR cs.CV
We present a new method for segmenting, and a new user interface for indexing and visualizing, the semantic content of extended instructional videos. Given a series of key frames from the video, we generate a condensed view of the data by clustering frames according to media type and visual similarities. Using various visual filters, key frames are first assigned a media type (board, class, computer, illustration, podium, and sheet). Key frames of media type board and sheet are then clustered based on contents via an algorithm with near-linear cost. A novel user interface, the result of two user studies, displays related topics using icons linked topologically, allowing users to quickly locate semantically related portions of the video. We analyze the accuracy of the segmentation tool on 17 instructional videos, each of which is from 75 to 150 minutes in duration (a total of 40 hours); the classification accuracy exceeds 96%.
cs/0302024
Analysis and Interface for Instructional Video
cs.IR cs.CV
We present a new method for segmenting, and a new user interface for indexing and visualizing, the semantic content of extended instructional videos. Using various visual filters, key frames are first assigned a media type (board, class, computer, illustration, podium, and sheet). Key frames of media type board and sheet are then clustered based on contents via an algorithm with near-linear cost. A novel user interface, the result of two user studies, displays related topics using icons linked topologically, allowing users to quickly locate semantically related portions of the video. We analyze the accuracy of the segmentation tool on 17 instructional videos, each of which is from 75 to 150 minutes in duration (a total of 40 hours); it exceeds 96%.
cs/0302029
Defeasible Logic Programming: An Argumentative Approach
cs.AI
The work reported here introduces Defeasible Logic Programming (DeLP), a formalism that combines results of Logic Programming and Defeasible Argumentation. DeLP provides the possibility of representing information in the form of weak rules in a declarative manner, and a defeasible argumentation inference mechanism for warranting the entailed conclusions. In DeLP an argumentation formalism will be used for deciding between contradictory goals. Queries will be supported by arguments that could be defeated by other arguments. A query q will succeed when there is an argument A for q that is warranted, ie, the argument A that supports q is found undefeated by a warrant procedure that implements a dialectical analysis. The defeasible argumentation basis of DeLP allows to build applications that deal with incomplete and contradictory information in dynamic domains. Thus, the resulting approach is suitable for representing agent's knowledge and for providing an argumentation based reasoning mechanism to agents.
cs/0302032
Empirical Methods for Compound Splitting
cs.CL
Compounded words are a challenge for NLP applications such as machine translation (MT). We introduce methods to learn splitting rules from monolingual and parallel corpora. We evaluate them against a gold standard and measure their impact on performance of statistical MT systems. Results show accuracy of 99.1% and performance gains for MT of 0.039 BLEU on a German-English noun phrase translation task.
cs/0302034
Interest Rate Model Calibration Using Semidefinite Programming
cs.CE
We show that, for the purpose of pricing Swaptions, the Swap rate and the corresponding Forward rates can be considered lognormal under a single martingale measure. Swaptions can then be priced as options on a basket of lognormal assets and an approximation formula is derived for such options. This formula is centered around a Black-Scholes price with an appropriate volatility, plus a correction term that can be interpreted as the expected tracking error. The calibration problem can then be solved very efficiently using semidefinite programming.
cs/0302035
Risk-Management Methods for the Libor Market Model Using Semidefinite Programming
cs.CE
When interest rate dynamics are described by the Libor Market Model as in BGM97, we show how some essential risk-management results can be obtained from the dual of the calibration program. In particular, if the objetive is to maximize another swaption's price, we show that the optimal dual variables describe a hedging portfolio in the sense of \cite{Avel96}. In the general case, the local sensitivity of the covariance matrix to all market movement scenarios can be directly computed from the optimal dual solution. We also show how semidefinite programming can be used to manage the Gamma exposure of a portfolio.