Navegación

Búsqueda

Búsqueda avanzada

Tuning Neural Networks in a Fuzzy Logic Programming Environment

Wide datasets are usually used for training and validating neural networks, which can be later tuned in order to correct their behaviors according to a few number of test cases proposed by users. In this paper we show how the FLOPER system developed in our research group is able to perform this last task after coding a neural network with a fuzzy logic language where program rules extend the classical notion of clause by including on their bodies both fuzzy connectives (useful for modeling activation functions of neurons) and truth degrees (associated to weights and bias in neural networks). We present an online tool which helps to select such operators and values in an automatic way, accomplishing with our recent technique for tuning this kind of fuzzy programs. Moreover, we provide some experimental results revealing that our tool generates the choices that better fit user’s preferences in a very efficient way, and producing relevant improvements on tuned neural networks.

Integrating WordNet into Bousi~Prolog (Work in Progress)

In this paper we provide techniques to integrate WordNet into a Fuzzy Logic Programming System. Because WordNet relates words but does not give graded information of the relation between them, we have implemented standard similarity measures and new directives that allow us to generate the proximity equations linking two words with an approximation degree. Proximity equations are the key syntactic structures that, in addition to a weak unification algorithm, make possible a flexible query answering process in this kind of programming languages.

An Efficient Proximity-based Unification Algorithm (Trabajo ya publicado)

Unification is a central concept in deductive systems based on the resolution principle. Recently, we introduced a new weak unification algorithm based on proximity relations (i.e., reflexive, symmetric, fuzzy binary relations). Proximity relations are able to manage vague or imprecise information and, in combination with the unification algorithm, allow certain forms of approximate reasoning in a logic programming framework. In this paper, we present a reformulation of the weak unification algorithm and an elaborated method to implement it efficiently. [This work has been accepted for its presentation at the 27th IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2018)]

An Online Tool for Unfolding Symbolic Fuzzy Logic Programs (Demostración)

In many declarative frameworks, unfolding is a very well-known semantics-preserving transformation technique based on the application of computational steps on the bodies of program rules for improving efficiency. In this paper we describe an online tool which allows us to unfold a symbolic extension of a modern fuzzy logic language where program rules can embed concrete and/or symbolic fuzzy connectives and truth degrees on their bodies. The system offers a comfortable interaction with users for unfolding symbolic programs and it also provides useful options to navigate along the sequence of unfolded programs. Finally, the symbolic unfolding transformation is connected with some fuzzy tuning techniques that we previously implemented on the same tool.

FuzzyDES: Fuzzifying DES (Trabajo en progreso)

This paper describes a system implementation of a fuzzy deductive database. Concepts supporting the fuzzy logic programming system BPL are translated into the deductive database system DES. We develop a version of fuzzy Datalog as its query language, where programs and queries are compiled to the DES core Datalog language. Weak unification and weak SLD resolution are adapted to this setting, and extended to allow rules with truth degree annotations. We provide a public implementation in Prolog which is open-source, multiplatform, portable, and in-memory. A database example for a recommender system is used to illustrate some of the features of the system.

Tuning Fuzzy Logic Programs with Symbolic Execution (Trabajo de alto nivel)

Fuzzy logic programming is a growing declarative paradigm aiming to integrate fuzzy logic into logic programming. One of the most difficult tasks when specifying a fuzzy logic program is determining the right weights for each rule, as well as the most appropriate fuzzy connectives and operators. In this paper, we introduce a symbolic extension of fuzzy logic programs in which some of these parameters can be left unknown, so that the user can easily see the impact of their possible values. Furthermore, given a number of test cases, the most appropriate values for these parameters can be automatically computed.

A fuzzy approach to cloud admission control for safe overbooking (High-level Work)

Cloud computing enables elasticity – rapid provisioning and deprovisioning of computational resources. Elasticity allows cloud users to quickly adapt resource allocation to meet changes in their workloads. For cloud providers, elasticity complicates capacity management as the amount of resources that can be requested by users is unknown and can vary significantly over time. Overbooking techniques allow providers to increase utilization of their data centers. For safe overbooking, cloud providers need admission control mechanisms to handle the tradeoff between increased utilization (and revenue), and risk of exhausting resources, potentially resulting in penalty fees and/or lost customers. We propose a flexible approach (implemented with fuzzy logic programming) to admission control and the associated risk estimation. Our measures exploit different fuzzy logic operators in order to model optimistic, realistic, and pessimistic behaviour under uncertainty. An experimental evaluation confirm that our fuzzy admission control approach can significantly increase resource utilization while minimizing the risk of exceeding the total available capacity.

Correctness of Incremental Model Synchronization with Triple Graph Grammars (High-level Work)

Cloud computing enables elasticity – rapid provisioning and deprovisioning of computational re-sources. Elasticity allows cloud users to quickly adapt resource allocation to meet changes in theirworkloads. For cloud providers, elasticity complicates capacity management as the amount of re-sources that can be requested by users is unknown and can vary significantly over time. Overbookingtechniques allow providers to increase utilization of their data centers. For safe overbooking, cloudproviders need admission control mechanisms to handle the tradeoff between increased utilization(and revenue), and risk of exhausting resources, potentially resulting in penalty fees and/or lost cus-tomers. We propose a flexible approach (implemented with fuzzy logic programming) to admissioncontrol and the associated risk estimation. Our measures exploit different fuzzy logic operators inorder to model optimistic, realistic, and pessimistic behaviour under uncertainty. An experimen-tal evaluation confirm that our fuzzy admission control approach can significantly increase resourceutilization while minimizing the risk of exceeding the total available capacity.

A Declarative Semantics for a Fuzzy Logic Language Managing Similarities and Truth Degrees

This work proposes a declarative semantics based on a fuzzy variant of the classical notion of least Herbrand model for the so-called FASILL language (acronym of “Fuzzy Aggregators and Similarity Into a Logic Language”) which has being recently designed and implemented in our research group for coping with implicit/explicit truth degree annotations, a great variety of connectives and unification by similarity.