Computational Intelligence

Published by Wiley
Online ISSN: 1467-8640
Print ISSN: 0824-7935
In symmetric cryptology (which is an essential part of modern computer security), the resistance to attacks depends critically on the nonlinearity properties of the Boolean functions describing cipher components like S-boxes. Some of the most effective methods known to generate functions that satisfy multiple criteria are based on evolutionary heuristics. In this paper, we improve on these algorithms by employing an adaptive strategy. Additionally, using recent improvements in the understanding of these combinatorial structures, we discover essential properties of the graph formed by affine equivalence classes of Boolean functions, which offers several advantages as a conceptual model for multiobjective seeking evolutionary heuristics. Finally, we propose the first major global cooperative effort to discover new bounds for cryptographic properties of Boolean functions.
We present a directed Markov random field (MRF) model, that combines n-gram models, probabilistic context free grammars (PC FGs) and probabilistic latent semantic analysis (PLSA), for the purpose of statistical language modeling. The composite directed MRF model has a potentially exponential number of loops and becomes a context sensitive grammar, nevertheless we are able to estimate its parameters in cubic time using an efficient modified ME method, the generalized inside-outside algorithm, which extends the inside-outside algorithm to incorporate the effects of the n-gram and PLSA language models.
This paper presents an overview and analysis of teaming in artificial neural systems (ANSs). It begins with a general introduction to neural networks and connectionist approaches to information processing. The basis for learning in ANSs is then described and compared with classical machine learning. While similar in some ways, ANS learning deviates from tradition in its dependence on the modification of individual weights to bring about changes in a knowledge representation distributed across connections in a network. This unique form of learning is analyzed from two aspects: the selection of an appropriate network architecture for representing the problem, and the choice of a suitable learning rule capable of reproducing the desired function within the given network. The various network architectures are classified, and then identified with explicit restrictions on the types of functions they are capable of representing. The learning rules, i.e., algorithms that specify how the network weights are modified, are similarly taxonomized and, where possible, the limitations inherent to specific classes of rules are outlined.
Even though considerable attention has been given to the polarity of words (positive and negative) and the creation of large polarity lexicons, research in emotion analysis has had to rely on limited and small emotion lexicons. In this paper we show how the combined strength and wisdom of the crowds can be used to generate a large, high-quality, word-emotion and word-polarity association lexicon quickly and inexpensively. We enumerate the challenges in emotion annotation in a crowdsourcing scenario and propose solutions to address them. Most notably, in addition to questions about emotions associated with terms, we show how the inclusion of a word choice question can discourage malicious data entry, help identify instances where the annotator may not be familiar with the target term (allowing us to reject such annotations), and help obtain annotations at sense level (rather than at word level). We conducted experiments on how to formulate the emotion-annotation questions, and show that asking if a term is associated with an emotion leads to markedly higher inter-annotator agreement than that obtained by asking if a term evokes an emotion.
This paper studies the properties of the continuous double auction trading mechanishm using an artificial market populated by heterogeneous computational agents. In particular, we investigate how changes in the population of traders and in market microstructure characteristics affect price dynamics, information dissemination and distribution of wealth across agents. In our computer simulated market only a small fraction of the population observe the risky asset’s fundamental value with noise, while the rest of agents try to forecast the asset’s price from past transaction data. In contrast to other artificial markets, we assume that the risky asset pays no dividend, so agents cannot learn from past transaction prices and subsequent dividend payments. We find that private information can effectively disseminate in the market unless market regulation prevents informed investors from short selling or borrowing the asset, and these investors do not constitute a critical mass. In such case, not only are markets less efficient informationally, but may even experience crashes and bubbles. Finally, increased informational efficiency has a negative impact on informed agents’ trading profits and a positive impact on artificial intelligent agents’ profits.
Many data mining and data analysis techniques operate on dense matrices or complete tables of data. Real-world data sets, however, often contain unknown values. Even many classification algorithms that are designed to operate with missing values still exhibit deteriorated accuracy. One approach to handling missing values is to fill in (impute) the missing values. In this paper, we present a technique for unsupervised learning called Unsupervised Backpropagation (UBP), which trains a multi-layer perceptron to fit to the manifold sampled by a set of observed point-vectors. We evaluate UBP with the task of imputing missing values in datasets, and show that UBP is able to predict missing values with significantly lower sum-squared error than other collaborative filtering and imputation techniques. We also demonstrate with 24 datasets and 9 supervised learning algorithms that classification accuracy is usually higher when randomly-withheld values are imputed using UBP, rather than with other methods.
Not all instances in a data set are equally beneficial for inferring a model of the data. Some instances (such as outliers) are detrimental to inferring a model of the data. Several machine learning techniques treat instances in a data set differently during training such as curriculum learning, filtering, and boosting. However, an automated method for determining how beneficial an instance is for inferring a model of the data does not exist. In this paper, we present an automated method that orders the instances in a data set by complexity based on the their likelihood of being misclassified (instance hardness). The underlying assumption of this method is that instances with a high likelihood of being misclassified represent more complex concepts in a data set. Ordering the instances in a data set allows a learning algorithm to focus on the most beneficial instances and ignore the detrimental ones. We compare ordering the instances in a data set in curriculum learning, filtering and boosting. We find that ordering the instances significantly increases classification accuracy and that filtering has the largest impact on classification accuracy. On a set of 52 data sets, ordering the instances increases the average accuracy from 81% to 84%.
This paper proposes an evolutionary approach for discovering difference in the usage of words to facilitate collaboration among people. When people try to communicate their concepts with words, the difference in the meaning and usage of words can lead to misunderstanding in communication, which can hinder their collaboration. In our approach each granule of knowledge in classification from users is structured into a decision tree so that difference in the usage of word can be discovered as the difference in the structure of tree. By treating each granule of classification knowledge as an individual in Genetic Algorithm (GA), evolution is carried out with respect to the classification efficiency of each individual and diversity as a population so that difference in the usage of words will emerge as the difference in the structure of decision tree. Experiments were carried out on motor diagnosis cases with artificially encoded difference in the usage of words and the result shows the effectiveness of our evolutionary approach. Keywordsusage of words-classification-decision tree-evolutionary approach
Cheeseman responds to responses to his previous essay 'An inquiry into computer understanding'. It is concluded that probability (not logic) is the best way to represent and reason about common sense in constructing an AI system.
This paper is about reducing influence diagram (ID) evaluation into Bayesian network (BN) inference problems that are as easy to solve as possible. Such reduction is interesting because it enables one to readily use one's favorite BN inference algorithm to efficiently evaluate IDs. Two such reduction methods have been proposed previously (Cooper 1988; Shachter and Peot 1992). This paper proposes a new method. The BN inference problems induced by the new method are much easier to solve than those induced by the two previous methods.
Current research on qualitative spatial representation and reasoning mainly focuses on one single aspect of space. In real world applications, however, multiple spatial aspects are often involved simultaneously. This paper investigates problems arising in reasoning with combined topological and directional information. We use the RCC8 algebra and the Rectangle Algebra (RA) for expressing topological and directional information respectively. We give examples to show that the bipath-consistency algorithm BIPATH is incomplete for solving even basic RCC8 and RA constraints. If topological constraints are taken from some maximal tractable subclasses of RCC8, and directional constraints are taken from a subalgebra, termed DIR49, of RA, then we show that BIPATH is able to separate topological constraints from directional ones. This means, given a set of hybrid topological and directional constraints from the above subclasses of RCC8 and RA, we can transfer the joint satisfaction problem in polynomial time to two independent satisfaction problems in RCC8 and RA. For general RA constraints, we give a method to compute solutions that satisfy all topological constraints and approximately satisfy each RA constraint to any prescribed precision.
In many large scale distributed systems and on the web, agents need to interact with other unknown agents to carry out some tasks or transactions. The ability to reason about and assess the potential risks in carrying out such transactions is essential for providing a safe and reliable environment. A traditional approach to reason about the trustworthiness of a transaction is to determine the trustworthiness of the specific agent involved, derived from the history of its behavior. As a departure from such traditional trust models, we propose a generic, machine learning approach based trust framework where an agent uses its own previous transactions (with other agents) to build a knowledge base, and utilize this to assess the trustworthiness of a transaction based on associated features, which are capable of distinguishing successful transactions from unsuccessful ones. These features are harnessed using appropriate machine learning algorithms to extract relationships between the potential transaction and previous transactions. The trace driven experiments using real auction dataset show that this approach provides good accuracy and is highly efficient compared to other trust mechanisms, especially when historical information of the specific agent is rare, incomplete or inaccurate.
One of the most important problems in the application of knowledge discovery systems is the identification and subsequent updating of rules. Many applications require that the classification rules be derived from data representing exemplar occurrences of data patterns belonging to different classes. The problem of identifying such rules in data has been researched within the field of machine learning, and more recently in the context of rough set theory and knowledge discovery in databases. In this paper we present an incremental methodology for finding all maximally generalized rules and for adaptive modification of them when new data become available. The methodology is developed in the context of rough set theory and is based on the earlier idea of discernibility matrix introduced by Skowron.
This paper presents a knowledge-based learning method and reports on case studies in different domains. The method integrates abduction and explanation-based learning. Abduction provides an improved method for constructing explanations. The improvement enlarges the set of examples that can be explained so that one can learn from additional examples using traditional explanation-based macro learning. Abduction also provides a form of knowledge level learning. Descriptions of case studies show how to set up abduction engines for tasks in particular domains. The case studies involve over a hundred examples taken from diverse domains requiring logical, physical, and psychological knowledge and reasoning. The case studies are relevant to a wide range of practical tasks including natural language understanding and plan recognition; qualitative physical reasoning and postdiction; diagnosis and signal interpretation; and decision making under uncertainty. The descriptions of the case studies include an example, its explanation, and discussions of what is learned by macro-learning and by abductive inference. The paper discusses how to provide and represent the domain knowledge and meta-knowledge needed for abduction and search control. The main conclusion is that abductive inference is important for learning. Abduction and macro-learning are complementary and synergistic.
Although there are many arguments that logic is an appropriate tool for artificial intelligence, there has been a perceived problem with the monotonicity of classical logic. This paper elaborates on the idea that reasoning should be viewed as theory formation where logic tells us the consequences of our assumptions. The two activities of predicting what is expected to be true and explaining observations are considered in a simple theory formation framework. Properties of each activity are discussed, along with a number of proposals as to what should be predicted or accepted as reasonable explanations. An architecture is proposed to combine explanation and prediction into one coherent framework. Algorithms used to implement the system as well as examples from a running implementation are given. Bien que de nombreux spécialistes soient d'avis que la logique est un outil approprié a l'intelligence artificielle, un problème relit a la monotonicite de la logique classique a été constaté. Cet article traite du concept selon lequel le raisonnement devrait ětre perçu comme une formation de theories a l'intérieur desquelles la logique nous indique les conséquences de nos hypothéses. L'activité de prediction de ce qui devrait ětre vrai et celle d'explication des observations sont examinées à l'intérieur d'un cadre de formation de théories. Les propriétés de chacune des activités sont discutsés, ainsi qu'un nombre de propositions concernant ce qui devrait ětre prédit ou accepté comme explication raisonnable. Une architecture est proposée afin de combiner explication et prédiction en un seul cadre cohérent. Les algorithmes utilisés pour mettre en œuvre ce systéme ainsi que des exemples sont fournis.
This article presents our work on the effective implementation of abduction in temporal reasoning. This works builds on some results, both in the logic programming field and in the automated reasoning area. We have defined and implemented an abductive procedure, which is well adapted for temporal reasoning because it is based on a constrained resolution principle. Constrained resolution has two advantages for temporal reasoning: First, it allows us to deal efficiently with temporal ordering and equality predicates, which are otherwise too much trouble with classical resolution; second, it allows a restricted form of abduction where hypotheses are limited to ordering relationships. From the logic programming area, our work uses results and procedures developed by others in the abductive logic programming field. The procedure we define and implement in this work is relatively independent of the temporal formalism: It has been used with some reified temporal logics and with the event calculus. More generally it can be used on any point-based temporal formalism, provided that a correct and complete algorithm is available for checking the consistency of a set of temporal ordering relationships in this language.
We describe a domain-independent framework (KNAVE) specific to the task of interpretation, summarization, visualization, explanation, and interactive navigation in a context-sensitive manner through time-oriented raw data and the multiple levels of higher-level, interval-based concepts that can be abstracted from these data. The KNAVE domainindependent navigation operators access the domainspecific knowledge base, which is modeled by the formal ontology of the knowledge-based temporal-abstraction method; the method generates the temporal abstractions from the time-oriented database. Thus, domain-specific knowledge underlies the semantics of the domainindependent visualization and navigation processes. By accessing the domain-specific temporal-abstraction knowledge base and the domain-specific time-oriented database, the KNAVE modules enable users to query for domain-specific temporal abstractions and to change the focus of the visualization, thus reusing for a different task (visualization and navigation) the domain model that has been acquired from the domain experts for the purpose of temporal abstraction. Initial evaluation of the KNAVE prototype has been encouraging. The KNAVE methodology has potentially broad implications for tasks such as planning, monitoring, explanation, and data mining.
In this paper we describe the implementation and evaluation of the AbTweak planning system, a test bed for studying and teaching concepts in partial-order planning, abstraction, and search control. We start by extending the hierarchical, preconditionelimination abstraction of Abstrips to partial-order-based, least-commitment planners such as Tweak. The resulting system, AbTweak, is used to illustrate the advantages of using abstraction to improve the efficiency of search. We show that by protecting a subset of abstract conditions achieved so far, and by imposing a bias on search toward deeper levels in a hierarchy, planning efficiency can be greatly improved. Finally, we relate AbTweak to other planning systems Snlp, Alpine and Sipe by exploring their similarities and differences. Computer Science Department. Waterloo, Ont. Canada, N2L 3G1. Tel: (519)888-4716. Fax: (519)8851208. Email: y Department of Mathematics and Computer Science, 1700 Mishawaka Ave...
xactly what this means, but I don't identify it with "frivolous") or aesthetic. Next, I argue that we cannot have acceptance, at least not in terms Kyburg seems to want it. Finally, I suggest some ideas that would have to be addressed in order to obtain an account of acceptance 2 worth accepting. Unfortunately, Kyburg's proposed acceptance approach comes bundled with all sorts of other baggage, including his own scheme for probability based on interval measures derived from statistical reference classes, as well as the general idea of evidential probability. These are ideas that Kyburg has certainly given a fair run for their money, and it is only natural that his acceptance theory would incorporate them. But since the issues they address are somewhat orthogonal to acceptance per se, I am afraid they confound the main points. In my response I will attempt to focus on the essential stance on acceptance, and defer taking issue with other elements of the
Nonlinear dynamical systems are notoriously difficult to control. The Acrobot is an under-actuated double pendulum in a gravitational field. Under most driving schemes the Acrobot exhibits chaotic behavior. But with careful applications of energy it is possible to gradually pump the system so as to swing it over its supporting joint. This swing-up task is of current interest to control theory researchers. Conventional notions of AI planning are not easily extended to domains with interacting continuously varying quantities. Such continuous domains are often dynamic; important properties change over time even when no action is taken. Noise and error propagation can preclude accurately characterizing the effects of actions or predicting the trajectory of an undisturbed system through time. A plan must be a conditional action policy or a control strategy that carefully nudges the system as it strays from a desired course. Automatically generating such plans or action strategies is the subject of this research. An AI system successfully learns to perform the swing-up task using an approach called explanation-based control (EBC). The approach combines a plausible qualitative domain theory with empirical observation. Results are in some respects superior to the known control theory strategies. Of particular importance to AI is EBC's notion of a “plan” or “strategy” and its method for automatic synthesis. Experimental evidence confirms EBC's ability and generality.
In this article we present the design of an ACL for a dynamic system of agents. The ACL includes a set of conversation performatives extended with operations to register, create, and terminate agents. The main design goal at the agent–level is to provide only knowledge–level primitives that are well integrated with the dynamic nature of the system. This goal has been achieved by defining an anonymous interaction protocol which enables agents to request and supply knowledge without considering symbol–level issues concerning management of agent names, routing, and agent reachability. This anonymous interaction protocol exploits a distributed facilitator schema which is hidden at the agent–level and provides mechanisms for registering capabilities of agents and delivering requests according to the competence of agents. We present a formal specification of the ACL and of the underlying architecture, exploiting an algebra of actors, and illustrate it with the help of a graphical notation. This approach provides the basis for discussing dynamic primitives in ACL and for studying properties of dynamic multi agent systems, for example concerning the behavior of agents and the correctness of their conversation policies.
Kasimir is a case-based decision support system in the domain of breast cancer treatment. For this system, a problem is given by the description of a patient and a solution is a set of therapeutic decisions. Given a target problem, Kasimir provides several suggestions of solutions, based on several justified adaptations of source cases. Such adaptation processes are based on adaptation knowledge. The acquisition of this kind of knowledge from experts is presented in this paper. It is shown how the decomposition of adaptation processes by introduction of intermediate problems can highlight simple and generalizable adaptation steps. Moreover, some adaptation knowledge units that are generalized from those acquired for Kasimir are presented. This knowledge can be instantiated in other case-based decision support systems, in particular in medicine.
Knowledge engineering for planning is expensive and the resulting knowledge can be imperfect. To autonomously learn a plan operator definition from environmental feedback, our learning system WISER explores an instantiated literal space using a breadth-first search technique. Each node of the search tree represents a state, a unique subset of the instantiated literal space. A state at the root node is called a seed state. WISER can generate seed states with or without utilizing imperfect expert knowledge. WISER experiments with an operator at each node. The positive state, in which an operator can be successfully executed, constitutes initial preconditions of an operator. We analyze the number of required experiments as a function of the number of the missing preconditions in a seed state. We introduce a naive domain assumption to test only a subset of the exponential state space. Since breadth- rst search is expensive, WISER introduces two search techniques to reorder literals at each l...
In this study, the adaptive neural fuzzy inference system (ANFIS), a hybrid fuzzy neural network, is adopted to predict the actions of the investors (when and whether they buy or sell) in a stock market in anticipation of an event–changes in interest rate, announcement of its earnings by a major corporation in the industry, or the outcome of a political election for example. Generally, the model is relatively more successful in predicting when the investors take actions than what actions they take and the extent of their activities. The findings do demonstrate the learning and predicting potential of the ANFIS model in financial applications, but at the same time, suggest that some of the market behaviors are too complex to be predictable.
this paper is on an action representation formalism that encodes both linguistic and planning knowledge about actions, and that supports the interpretation of complex Natural Language instructions, and in particular, of instructions containing Purpose Clauses. The representation uses linguistically motivated primitives, derived from Jackendoff 's work on Conceptual Semantics, and is embedded in the description logic based system CLASSIC. I will first motivate the characteristics of the formalism as needed to understand Natural Language instructions. I will then describe the formalism itself, and I will argue that the integration of a linguistically motivated lexical semantics formalism and of a description logic based system is beneficial to both. Finally, I will show how the formalism is exploited by the algorithm that interprets Purpose Clauses. The output of the algorithm is used in the Animation from NL project, that has as its goal the automatic creation of animated task simulations
A clear understanding and formalization of actions is essential to computing, and especially so to reasoning about and constructing intelligent agents. Several approaches have been proposed over the years. However, most approaches concentrate on the causes and effects of actions, but do not give general characterizations of actions themselves. A useful formalization of actions would be based on a general, possibly nondiscrete, model of time that allows branching (to capture agents' choices). A desirable formalization would also allow actions to be of arbitrary duration and would permit multiple agents to act concurrently. We develop a branching-time framework that allows great flexibility in how time and action are modeled. We motivate and formalize several coherence constraints on our models, which capture some nice intuitions and validate some useful inferences relating actions with time. 1 Introduction Over the years, actions and time have garnered much research attent...
We use the idea that actions performed in a conversation become part of the common ground as the basis for a model of context that reconciles in a general and systematic fashion the differences between the theories of discourse context used for reference resolution, intention recognition, and dialogue management. We start from the treatment of anaphoric accessibility developed in DRT, and we show first how to obtain a discourse model that, while preserving DRT's basic ideas about referential accessibility, includes information about the occurrence of speech acts and their relations. Next, we show how the different kinds of `structure' that play a role in conversation--- discourse segmentation, turn-taking, and grounding---can be formulated in terms of information about speech acts, and use this same information as the basis for a model of the interpretation of fragmentary input. 1 Motivations Although the slogan `language is (joint) action' is accepted by almost everyone w...
Hypothetical reasoning about actions is the activity of preevaluating the effect of performing actions in a changing domain; this reasoning underlies applications of knowledge representation, such as planning and explanation generation. Action effects are often specified in the language of situation calculus, introduced by McCarthy and Hayes in 1969. More recently, the event calculus has been defined to describe actual actions, i.e., those that have occurred in the past, and their effects on the domain. Altough the two formalisms share the basic ontology of atomic actions and fluents, situation calculus cannot represent actual actions while event calculus cannot represent hypotethical actions. In this article, the language and the axioms of event calculus are extended to allow representing and reasoning about hypothetical actions, performed either at the present time or in the past, altough counterfactuals are not supported. Both event calculus and its extension are defined as logic programs so that theories are readily adaptable for Prolog query interpretation. For a reasonably large class of theories and queries, Prolog interpretation is shown to be sound and complete w.r.t. the main semantics for logic programs.
To plan means reasoning about possible actions, but a robot must also reason about actual events. This paper proposes a formal theory about actual and possible events. It presents a new modal logic as a notation for this theory and a technique for planning in the modal logic using a first-order theorem prover augmented with simple modal reasoning. This avoids the need for a general modal-logic theorem prover. Adding beliefs to this theory raises an interesting problem for which the paper offers a tentative solution. Planifier signifie raisonner sur des actions possible, mais un robot doit aussi raisonner sur des évènements reels. Cet article propose une théorie formelle traitant des évènements réels et possibles. II présente une nouvelle logique modale servant de notation pour cette théorie, et une technique de génération de plan en logique modale utilisant un démonstrateur de thdoreme du premier ordre augmente d'un raisonneur modal simple. Ceci permet d'eviter le recours a un demonstrateur de theoreme base sur la logique modale trop general. Le fait d'ajouter les croyances a cette theorie souleve un probème intéressant pour lequel cet article propose une solution provisoire. Mots clés: génération de plan, mondes possibles, logique modale, croyance, raisonnement non-monotone.
Artificial immune systems (AIS) are computational systems inspired by the principles and processes of the vertebrate immune system. The AIS-based algorithms typically exploit the immune system's characteristics of learning and adaptability to solve some complicated problems. Although, several AIS-based algorithms have proposed to solve multi-objective optimization problems (MOPs), little focus have been placed on the issues that adaptively use the online discovered solutions. Here, we proposed an adaptive selection scheme and an adaptive ranks clone scheme by the online discovered solutions in different ranks. Accordingly, the dynamic information of the online antibody population is efficiently exploited, which is beneficial to the search process. Furthermore, it has been widely approved that one-off deletion could not obtain excellent diversity in the final population; therefore, a k-nearest neighbor list (where k is the number of objectives) is established and maintained to eliminate the solutions in the archive population. The k-nearest neighbors of each antibody are founded and stored in a list memory. Once an antibody with minimal product of k-nearest neighbors is deleted, the neighborhood relations of the remaining antibodies in the list memory are updated. Finally, the proposed algorithm is tested on 10 well-known and frequently used multi-objective problems and two many-objective problems with 4, 6, and 8 objectives. Compared with five other state-of-the-art multi-objective algorithms, namely NSGA-II, SPEA2, IBEA, HYPE, and NNIA, our method achieves comparable results in terms of convergence, diversity metrics, and computational time.
The formalism of synchronous tree-adjoining grammars, a variant of standard tree-adjoining grammars (TAG), was intended to allow the use of TAGs for language transduction in addition to language specification. In previous work, the definition of the transduction relation defined by a synchronous TAG was given by appeal to an iterative rewriting process. The rewriting definition of derivation is problematic in that it greatly extends the expressivity of the formalism and makes the design of parsing algorithms difficult if not impossible. We introduce a simple, natural definition of synchronous tree-adjoining derivation, based on isomorphisms between standard tree-adjoining derivations, that avoids the expressivity and implementability problems of the original rewriting definition. The decrease in expressivity, which would otherwise make the method unusable, is offset by the incorporation of an alternative definition of standard tree-adjoining derivation, previously proposed for completely separate reasons, thereby making it practical to entertain using the natural definition of synchronous derivation. Nonetheless, some remaining problematic cases call for yel more flexibility in the definition; the isomorphism requirement may have to be relaxed. It remains for future research to rune the exact requirements on the allowable mappings.
This paper describes a natural language generation system developed for a spoken language translation system. Our system employs Feature-structure-based Tree Adjoining Grammar (FTAG) for generation knowledge representation. Each elementary tree of our grammar is paired with a semantic feature structure which is consistent with semantics defined in Head-driven Phrase Structure Grammar. Feature structures attached to nodes of elementary trees are unrestricted. Thus our formalism allows HPSG style phrase structure description as well as TAG style description. The advantage of our generation knowledge representation is the ability of incorporating HPSG style “core” grammar and TAG style case-based grammar. The system generates a syntactic tree by combining elementary trees so as to satisfy an input semantic structure. The generation algorithm is an application of a semantic head-driven generation. To carry out an adjoining operation. an elementary tree with an adjunction node is dynamically split at the adjunction node during generation.
In many head-final languages such as German, Hindi, Japanese, and Korean, but also in some other languages such as Russian, arguments of a verb can occur in any order. Furthermore, arguments can occur outside of their clause ("long-distance scrambling"). Long-distance scrambling presents a challenge both to linguistic theory and to formal frameworks for linguistic description because it is very unconstrained: in given sentence, there is no bound on the number of elements that can scrambled nor on the distance over which each element can scramble. We discuss two formal frameworks related to Tree Adjoining Grammar. First, we show how scrambling facts from Korean can be handled by non-local multi-component TAG (MC-TAG). Then, we argue that overt wh- movement in German makes this analysis unattractive, and suggest a new version of MC-TAG, called V-TAG, which can handle both Korean and German word order variation. Interestingly, this new version has more attractive computational properties ...
Partial descriptions are sets of constraints which do not necessarily determine a unique object. Two related issues arise when objects are specified by partial descriptions. The first asks whether a given description is satisfied by any object. The second seeks to produce, from a description, an appropriate representative of those objects which satisfy it. We explore these issues in the context of two recent proposals utilizing partial descriptions of trees in the area of Tree-Adjoining Grammars. In this context, what counts as an appropriate representative of the set of trees which satisfy a description is a tree in that set which is minimal in the sense that every other tree in the set can be derived from it by adjunction. We call a partial description for which there is a unique such minimal tree a quasi-tree. We formalize the notions of partial descriptions of trees and of quasi-trees, and, using these, provide mechanisms which resolve any description of trees into an equivalent s...
Tree-adjoining grammars (TAG) have been proposed as a formalism for generation based on the intuition that the extended domain of syntactic locality that TAGs provide should aid in localizing semantic dependencies as well, in turn serving as an aid to generation from semantic representations. We demonstrate that this intuition can be made concrete by using the formalism of synchronous tree-adjoining grammars. The use of synchronous TAGs for generation provides solutions to several problems with previous approaches to TAG generation. Furthermore, the semantic monotonicity requirement previously advocated for generation grammars as a computational aid is seen to be an inherent property of synchronous TAGs. Les grammaires ?arbres adjoints possèdent un domaine de localité syntaxique étendu. À priori cette měme proprieté permet aussi de localiser certaines dépendences sémantiques. Basées sur cette intuition, les grammaires ?arbres adjoints ont été proposées comme formalisme pour la génération de texte. Dans ce papier, les auteurs montrent comment les grammaires ?arbres adjoints synchrones concrétised cette intuition de maniere précise. L'utilisation des grammaires ?arbres adjoints synchrones résout plusieurs problèmes liés à des approches précédentes pour la génération de texte. De plus, la monotonicite sémantique, propriété souvent préférée pour la génération de texte, est inhèrente aux grammaires ?arbres adjoints synchrones.
This paper presents an algorithm (a parser) for analyzing sentences according to grammatical constraints expressed in the framework of lexicalized tree-adjoining grammar. For the current grammars of English, the algorithm behaves much better and requires much less time than its worst-case complexity. The main objective of this work is to design a practical parser whose average-case complexity is much superior to its worst case. Most of the previous methods always required the worst-case complexity. The algorithm can be used in two modes. As a recognizer it outputs whether the input sentence is grammatically correct or not. As a parser it outputs a detailed analysis of the grammatically correct sentences. As sentences are read from left to right, information about possible continuations of the sentence is computed. In this sense, the algorithm is called a predictive left to right parser. This feature reduces the average time required to process a given sentence. In the worst case, the parser requires an amount of time proportional to G2n6 for a sentence of n words and for a lexicalized tree-adjoining grammar of size G. The worst-case complexity is only reached with pathological (not naturally occurring) grammars and inputs.
In this paper we provide an implementation strategy to map a functional specification of an utterance into a syntactically well‐formed sentence. We do this by integrating the functional and the syntactic perspectives on language, which we take to be exemplified by systemic grammars and tree adjoining grammars (TAGs) respectively. From systemic grammars we borrow the use of networks of choices to classify the set of possible constructions. The choices expressed in an input are mapped by our generator to a syntactic structure as defined by a TAG. We argue that the TAG structures can be appropriate structural units of realization in an implementation of a generator based on systemic grammar and also that a systemic grammar provides an effective means of deciding between various syntactic possibilities expressed in a TAG grammar. We have developed a generation strategy which takes advantage of what both paradigms offer to generation, without compromising either.
Negotiations are very important in a multi-agent environment, particularly, in an environment where there are conflicts between the agents, and cooperation would be beneficial. We have developed a general structure for a Negotiating Automated Agent that consists of five modules: a Prime Minister, a Ministry of Defense, a Foreign Office, a Headquarters and Intelligence. These modules are implemented using a dynamic set of local-agents belonging to the different modules. We used this structure to develop a Diplomacy player, Diplomat. Playing Diplomacy involves a certain amount of technical skills as in other board games, but the capacity to negotiate, explain, convince, promise, keep promises or break them, is an essential ingredient in good play. Diplomat was evaluated and consistently played better than human players. Key Words: Automated Negotiations, Multi-Agent Environment, Game Playing, Localagents, Diplomacy. Subject Category: Cognitive Science, Knowledge Representati...
Automated negotiation is widely applied in various domains. However, the development of such systems is a complex knowledge and software engineering task. So, a methodology there will be helpful. Unfortunately, none of existing methodologies can offer sufficient, detailed support for such system development. To remove this limitation, this paper develops a new methodology made up of (1) a generic framework (architectural pattern) for the main task, and (2) a library of modular and reusable design pattern (templates) of subtasks. Thus, it is much easier to build a negotiating agent by assembling these standardized components rather than reinventing the wheel each time. Moreover, because these patterns are identified from a wide variety of existing negotiating agents (especially high impact ones), they can also improve the quality of the final systems developed. In addition, our methodology reveals what types of domain knowledge need to be input into the negotiating agents. This in turn provides a basis for developing techniques to acquire the domain knowledge from human users. This is important because negotiation agents act faithfully on the behalf of their human users and thus the relevant domain knowledge must be acquired from the human users. Finally, our methodology is validated with one high impact system.
This paper presents a formal framework and outlines a method that autonomous agents can use to negotiate the semantics of their communication language at run-time. Such an ability is needed in open multi-agent systems so that agents can ensure they understand the implications of the utterances that are being made and so that they can tailor the meaning of the primitives to best fit their prevailing circumstances. To this end, the semantic space framework provides a systematic means of classifying the primitives along multiple relevant dimensions. This classification can then be used by the agents to structure their negotiation (or semantic fixing) process so that they converge to the mutually agreeable semantics that are necessary for coherent social interactions
As they gain expertise in problem solving, people increasingly rely on patterns and spatiallyoriented reasoning. This paper describes an associative visual pattern classifier and the automated acquisition of new, spatially-oriented reasoning agents that simulate such behavior. They are incorporated into a multi-agent game-learning program whose architecture robustly combines agents with conflicting perspectives. When tested on three games, the visual pattern classifier learns meaningful patterns, and the pattern-based, spatially-oriented agents generalized from these patterns are generally correct. The accuracy of the contribution of each of the newly created agents to the decision-making process is measured against an expert opponent, and a perceptron-like algorithm is used to learn game-specific weights for these agents. Much of the knowledge encapsulated by the new agents was previously inexpressible in the program's representation and in some cases is not readily deducible from the...
Agent technology is an exciting and important new way to create complex software systems. Agents blend many of the traditional properties of AI programs—knowledge–level reasoning, flexibility, proactiveness, goal–directedness, and so forth—with insights gained from distributed software engineering, machine learning, negotiation and teamwork theory, and the social sciences. An important part of the agent approach is the principle that agents (like humans) can function more effectively in groups that are characterized by cooperation and division of labor. Agent programs are designed to autonomously collaborate with each other in order to satisfy both their internal goals and the shared external demands generated by virtue of their participation in agent societies. This type of collaboration depends on a sophisticated system of inter–agent communication. The assumption that inter–agent communication is best handled through the explicit use of an agent communication language (ACL) underlies each of the articles in this special issue. In this introductory article, we will supply a brief background and introduction to the main topics in agent communication.
In this paper, we propose a strategic-negotiation model which enables self-motivated rational agents to share resources. The strategic-negotiation model takes the passage of time during the negotiation process itself into account. The model considers bilateral negotiations in situations characterized by complete information, in which one agent loses over time while the other gains over time. Using this negotiation mechanism, autonomous agents apply simple and stable negotiation strategies that result in efficient agreements without delays, even when there are dynamic changes in the environment.
This paper proposes a replanning mechanism for deliberative agents as a new approach to tackling the frame problem. We propose a beliefs desires and intentions (BDI) agent architecture using a case-based planning (CBP) mechanism for reasoning. We discuss the characteristics of the problems faced with planning where constraint satisfaction problems (CSP) resources are limited and formulate, through variation techniques, a reasoning model agent to resolve them. The design of the agent proposed, named MRP-Ag (most-replanable agent), will be evaluated in different environments using a series of simulation experiments, comparing it with others such as E-Ag (Efficient Agent) and O-Ag (Optimum Agent). Last, the most important results will be summarized, and the notion of an adaptable agent will be introduced.
Coalition formation is an important mechanism for cooperation in multiagent systems. In this paper we address the problem of coalition formation among self-interested agents in superadditive task-oriented domains. We assume that each agent has some “structure,” i.e., that it can be described by the values taken by a set of m nonnegative attributes that represent the resources w each agent is endowed with. By defining the coalitional value as a function V of w, we prove a sufficient condition for the existence of a stable payment configuration—in the sense of the core—in terms of certain properties of V. We apply these ideas to a simple case that can be described by a linear program and show that it is possible to compute for it—in polynomial time—an optimal task allocation and a stable payment configuration.
We define and study social constraints for rational agents. Our work is complementary to work on mechanism design in economics and Distributed Artificial Intelligence, as well as to work on artificial social systems. In our setting agents are rational but obey social laws that are imposed by the system's designer. Agents can be obliged to obey some social constraints, but not any constraint can serve as part of the social law. The main theme of our work is the study of settings where there are restrictions on the constraints that can serve as social laws. In such settings the designer should find social laws that can be imposed on the agents, and that will lead rational agents to satisfactory behavior. Our study is carried out in the context of zero-sum and general-sum games (with complete and with incomplete information) in extensive form.
This paper presents a framework that captures how the social nature of agents that are situated in a multi-agent environment impacts upon their individual mental states. Roles and social relationships provide an abstraction upon which we develop the notion of social mental shaping. This allows us to extend the standard Belief-DesireIntention model to account for how common social phenomena (e.g. cooperation, collaborative problem-solving and negotiation) can be integrated into a unified theoretical perspective that reflects a fully explicated model of the autonomous agent's mental state. Keywords: Multi-agent systems, agent interactions, BDI models, social influence. 3 1.
The content of real-world databases, knowledge bases, database models, and formal specifications is often highly redundant and needs to be aggregated before these representations can be successfully paraphrased into natural language. To generate natural language from these representations, a number of processes must be carried out, one of which is sentence planning where the task of aggregation is carried out. Aggregation, which has been called ellipsis or coordination in Linguistics, is the process that removes redundancies during generation of a natural language discourse, without losing any information.The article describes a set of corpus studies that focus on aggregation, provides a set of aggregation rules, and finally, shows how these rules are implemented in a couple of prototype systems. We develop further the concept of aggregation and discuss it in connection with the growing literature on the subject. This work offers a new tool for the sentence planning phase of natural language generation systems.
While there has been recent interest in research on planning and reasoning about actions, nearly all research results have been theoretical. We know of no previous examples of a planning system that has made a significant impact on a problem of practical importance. One of the primary goals during the development of the SIPE-2 planning system has been the balancing of efficiency with expressiveness and flexibility. With a major new extension, SIPE-2 has begun to address practical problems. This paper describes this new extension and the new applications of the planner. One of these applications is the problem of producing products from raw materials on process lines under production and resource constraints. This is a problem of commercial importance and SiPE-2's application to it is described in some detail. Bien que Ton ait constaté récemment un intérět pour la planification et le raisonnement à propos des actions, presque tous les résuhats des recherches sont théoriques. Nous ne connaissons pas d'exemple de système de planification qui ait eu une influence majeure dans la résolution d'un problème de nature pratique. L'un des prihcipaux objectifs de l'élaboration du système de planification SIPE-2 a été la recherche d'un équilibfe entre l'efficacité, l'expressivité et la flexibilityé. La capacityé du système SIPE-2 a été augmentée considérablement afin qu'il puisse s'attaquer à la résolution de problèmes de nature pratique. Cet article traite de 1'extension du système et des nouvelles applications du planificateur. L'une de ces nouvelles applications est relative au problème de la production à la chaine de biens à partir de matières premières, en tenant compte de contraintes de production et de ressources. Ce problème est d'une importance commerciale et l'application du système SIPE-2 à sa résolution est decrite avec plus ou moins de détails.
Top-cited authors
Saif M. Mohammad
  • National Research Council Canada
Peter David Turney
  • Ronin Institute
David Israel
  • SRI International
Thomas Dean
  • Brown University
Henri Prade
  • Paul Sabatier University - Toulouse III