NAFIPS’2016

Skyline of El Paso Texas

October 31 - November 4, 2016

El Paso, Texas

As the premier fuzzy society in North America established in 1981, NAFIPS's purpose is to help guide and encourage the development of fuzzy sets and related technologies for the benefit of mankind.

NAFIPS 2016 -- PLENARY SPEAKERS

List of plenary speakers and talks' abstracts:

  • Samir Abou-Samra, DigiPen Institute of Technology, USA
    Title: Calibrating a Video Game Using Fuzzy Logic
    Abstract: Fuzzy logic is generally applied in video games to simulate the artificial intelligence behavior of Non Playable Characters (NPC). This talk will illustrate an elevated application in video games: illusion of suspension of disbelief. When taken in the context of video games being a part of the entertainment industry, a well-calibrated game utilizing this sophisticated method produces most commercial success in the market. The recipe used to keep gamers entertained by feeding their ego and satisfying their greed is entirely dependent on this secret ingredient: fuzzy logic.

  • [Canceled] Ildar Batyrshin, Centro de Investigación en Computación, Instituto Politechnico Nacional, Mexico City, Mexico
    Title: Fuzzy Logic Based Analysis of Similarity and Association Measures
    Abstract: Similarity and association measures play an important role in data mining and data analysis, in clustering and pattern classification, in machine learning and decision making, in time series analysis and forecasting, in social network analysis and recommender systems, in bioinformatics, in text processing, in medical applications etc. In this talk, it is argued that the fuzzy logic in the broad sense can serve as a basis for analysis of similarity and association measures in general and particularly in specific domains. The properties and the methods of construction of different similarity and association measures are considered.

  • Piero Bonissone, Piero P Bonissone Analytics, LLC, CEO, USA
    Title: Analytics for Industrial Internet Applications
    Abstract: The Industrial Internet is the third disruptive wave, after the Industrial and the Internet revolutions. It is transforming our industries, just like the Internet revolution transformed our commerce. In this new context, we face a combination of hyper-connected intelligent machines, interacting with other machines and people, and generating large amounts data that need to be analyzed by descriptive, predictive, and prescriptive models. As a result, we see the resurgence of analytics as a key differentiator for creating new services, the emergence of cloud computing as an enabling technology for service delivery, and the growth of crowdsourcing as a new phenomenon in which people play critical roles in creating information and shaping decisions in a variety of problems. We explore the intersection of these three concepts from the perspective of a machine-learning researcher and show how his job and roles have evolved over time. In the past, analytic model creation was an artisanal process, as models were handcrafted by experienced, knowledgeable model-builders. More recently, the use of meta-heuristics, such as evolutionary algorithms, has provided us with limited levels of automation in model building and maintenance. In the short future, we expect data-driven analytic models to become a commodity. We envision having access to a large number of data-driven models, obtained by a combination of crowdsourcing, cloud-based evolutionary algorithms, outsourcing, in-house development, and legacy models. In this context, the critical issue will be model ensemble selection and fusion, rather than model generation. First, we will review the application of data-driven analytic models to assets diagnostics and prognostics, such as aircraft engines, medical imaging devices, and locomotives. We will cover a case study on prediction of remaining useful life for each unit in a fleet of locomotives using fuzzy models. Then we will explore the evolution of analytic models with the advent of cloud computing, and propose the use of customized model ensembles on demand, inspired by Lazy Learning. This approach is agnostic with respect to the origin of the models, making it scalable and suitable for a variety of applications. We successfully tested this approach in a regression problem for a power plant management application, using two different sources of models: bootstrapped neural networks, and GP-created symbolic regression models evolved in the cloud. We will also present results on the fusion of models for FlyQuest, a GE-sponsored Kaggle competition in which we crowd-sourced the generation of models predicting the estimated runway and gateway arrival (ERA, EGA) over a month of US flights. Finally, we will explore research trends, challenges and opportunities for Machine Learning techniques in this emerging context of big data and cloud computing.

  • Martine De Cock, Institute of Technology of the University of Washington Tacoma & Ghent University, Belgium
    Title: Fuzzy Answer Set Programming: from Theory to Practice -- full presentation
    Abstract: Answer Set Programming (ASP) is a popular declarative programming paradigm which allows for an easy and intuitive encoding of many combinatorial search and optimisation problems. It enables non-monotonic reasoning by virtue of a negation-as-failure operator with a purely declarative semantics. An ASP program describes a problem as a set of rules. This set of rules is fed to an answer set solver that finds stable models (i.e. answer sets) which correspond to the solutions of the considered problem. The availability of fast and efficient solvers for ASP has paved the way for ASP applications in many fields.
    Despite its flexibility and expressive power, ASP lacks the ability to directly encode problems in continuous domains, because it is based on Boolean logic. Fuzzy Answer Set Programming (FASP) is an extension of ASP that uses multi-valued semantics for evaluating propositions. In this talk we cover our journey from the theoretical work on (1) defining the syntax and semantics of FASP and (2) pinning down the computational complexity of solving different kinds of FASP-programs, to the more practical aspects: (3) the development of a FASP solver and (4) its use for simulating the dynamics of gene regulatory networks with multi-valued activation levels.

  • Victor Raskin, Purdue University, USA
    Title: Fuzzy Linguistic Semantics of/for/in Cybersecurity

  • Christian Servin, El Paso Community College, El Paso, Texas, USA
    Title: Fuzzy Techniques Help on All Stages of Knowledge and Data Processing: Applied Case Studies -- full presentation
    Abstract: Traditionally, fuzzy techniques have been mostly used in one _specific_ task: to translate expert knowledge into precise terms. In this talk, we show, on application examples, that fuzzy techniques are very helpful on _all_ the stages of knowledge and data processing: from gauging measurement accuracy and spatial resolution to testing hypothesis to estimating the values of the desired quantities.

  • Dongrui Wu, DataNova LLC, USA
    Title: Fuzzy logic and machine learning for brain-computer interface (BCI)
    Abstract: Brain computer interfaces (BCIs) have attracted rapidly increasing research interest in the last decade, thanks to recent advances in neurosciences, wearable/mobile biosensors, and analytics. However, there are many challenges in their transition from laboratory settings to real-life applications, including the reliability and convenience of the sensing hardware, the availability of high-performance and robust algorithms for signal analysis and interpretation, and fundamental advances in automated reasoning that effectively handle individual differences and nonstationarity. This talk focuses on the last challenge, more specifically, how to generalize a BCI algorithm to a new subject, with zero or very little subject-specific calibration data. It will show that by combining fuzzy logic with machine learning approaches, particularly active learning, deep learning, and transfer learning, we can significantly reduce the amount of subject-specific calibration data required in BCI, for both online and offline applications, and both classification and regression problems.

Conference location

  • The University of Texas at El Paso
    Union Building - East
    3rd floor
    Rooms: Tomas Rivera Conference Center, University Suite, Templeton Suite, and Ray room

  • See map of UTEP
  • See map of Union East's 3rd floor

NAFIPS 2016 Program

Contact Information

Important Dates

  • 03/07/2016: Special session proposals
  • 06/14/2016: Papers due (extended deadline)
  • 10/31-11/04/2016: Conference

About the Location

About El Paso

Downtown El Paso

Discover a city that stretches the imagination - El Paso, Texas' westernmost city. A city of nearly three-quarters of a million people, which sprawls across hundreds of square miles of desert and rambling foothills. More about El Paso

About UTEP

University of Texas at El Paso

Just past its first centennial celebration, UTEP is the second oldest academic component of the University of Texas system. It has over 20,000 students. It is a beautiful, friendly, and newly remodeled campus with a harmonious and pleasing Bhutanese architecture.

Where to Stay

Stay In El Paso Screenshot

Although the most convenient hotel is probably the on-campus Hilton Garden Inn, many other options are available in the campus' neighborhood, and can be checked on the website of the El Paso Hotel Motel Association.

What to do

The Sunbowl at UTEP

El Paso is an intriguing city worth visiting. Its mountains, the Franklin Mountains, southmost part of the Rockies, are surely a place to visit. An easy way to get a nice view of the city is by taking the Wyler aerial tramway. Surrounding places of interest also include the New Mexico sites: La Mesilla, Carlsbad caverns, White Sands National Monument.