Plenary Talks – Abstracts

MONDAY

Prof. Cristovam Buarque (University of Brasilia / Brazilian Senate, Brazil)

“Emergência Tardia”

O surgimento dos países emergentes é um dos fenômenos mais importantes da passagem dos séculos. Para surpresa geral, surgiu um grupo de países que se aproximam em riqueza e influência do seleto grupo de países ricos. Dentre estes países emergentes, um subgrupo especial realiza um dos mais importantes fenômenos da geopolítica e da diplomacia destes tempos, ao se constituírem como os BRICS. Tudo indicava que do ponto de vista econômico o mundo caminhava para o constante enriquecimento global que absorveria os países do BRICS, recém emergidos na riqueza.

De repente porém começa-se a perceber que os emergentes chegaram tardiamente a um conceito de riqueza que se esgota. É como se fossem emergentes tardios. Chegam à superfície do mar no momento em que falta oxigênio no ar. O modelo de progresso que ao longo de décadas não levou em conta os limites ecológicos, nem os limites financeiros e fiscais, entra em crise, buscando um novo tipo de progresso, um novo modelo de desenvolvimento que vá além do crescimento econômico. Formular este novo modelo é um desafio dos BRICS.

Armando Monteiro Neto (Brazilian Senate, Brazil)

“Momento Econômico e papel da Academia”

Na atual sociedade do conhecimento, a inovação e a tecnologia se apresentam como principais impulsionadores do crescimento econômico. Estudos apontam que mais da metade da riqueza mundial é gerada pelo conhecimento, superando os fatores tradicionais (recursos naturais, capital e trabalho). Isso exige do setor produtivo maior capacidade de inovação, como determinante primordial da competitividade. Nesse contexto, as universidades devem assumir o duplo papel de gerar o conhecimento básico e de criar mecanismos eficazes de transferência de tecnologia. O desafio dos países emergentes, como os BRICS, passa pelo fortalecimento de um ambiente institucional que favoreça a integração entre o setor produtivo, em especial a indústria, e as universidades no desenvolvimento de competências para inovação.

Prof. Igor Aleksander (Imperial College London, England)

“The Conscious Ghost In The Neural Machine”

What progress has been made in linking the known analytic properties of neural systems to the first person sensations experienced by living entities? I suggest that the drive towards ever-greater autonomy in artificial  systems (e.g. the ‘Curiosity’ rover) does require an ‘artificial’ form of consciousness. I further argue that systems that achieving the needed autonomy will need a structure of internal states that represent the sense of reality that philosophers call ‘phenomenal’. I describe three major formal properties that an internal state must have in order to provide the artificial object with its phenomenal reality: (a) internal states that reasonably represent and match the objects of sensory perception, (b) the necessary uniqueness that such states need to be informative about the world and (c) the necessary integration that makes such states irreducible. I refer to our own work [Aleksander and Morton: “Aristotle’s Laptop: The Discovery of our Informational Minds”, WSPC, 2012] and that of Tononi [ Biological Bulletin 215,  pp 216-242 (2008)] to show that these three properties translate into connectivity requirements in neural nets, providing design criteria for ‘the conscious ghost in the neural machine’. I shall end with the challenges that shape future work.

Prof. Derong Liu (Institute of Automation – Chinese Academy of Sciences, China)

“Self-Learning Control of Nonlinear Systems based on Iterative Adaptive Dynamic”

The optimal control of nonlinear systems often requires solving the nonlinear Hamilton-Jacobi-Bellman (HJB) equation instead of the Riccati equation as in the linear case. The discrete-time HJB (DTHJB) equation is more difficult to work with than the Riccati equation because it involves solving nonlinear partial difference equations. Though dynamic programming has been a useful computational technique in solving optimal control problems for many years, it is often computationally untenable to run it to obtain the optimal solution, due to the backward numerical process required for its solutions, i.e., the well-known “curse of dimensionality”. A self-learning control scheme for unknown nonlinear discrete-time systems is developed for this purpose. An iterative adaptive dynamic programming algorithm via globalized dual heuristic programming technique is developed to obtain the optimal controller with convergence analysis. Neural networks are used as parametric structures to facilitate the implementation of the iterative algorithm, which will approximate at each iteration the cost function, the optimal control law, and the unknown nonlinear system, respectively. Simulation examples are provided to verify the effectiveness of the present self-learning control approach.

Prof.  G. Kumar Venayagamoorthy (Clemson University, USA)

“Computational Intelligence Methods for Real-Time Operation of Smart Grids”

With the emerging innovations to the electricity infrastructure (referred to as the smart grid), high levels of penetration of renewable energy, and an emphasis on competitive pricing, it will become necessary to optimize the safety margins presently allowed, and use existing equipment as optimally as possible. Maintaining reliable service and implementing emergency defense plans during major unintended disturbances and intended attacks is critical with the growth of the electric power network and its information infrastructure. The development of reliable and scalable intelligent monitoring and control algorithms, and situational intelligence (beyond situational awareness (SA)) technologies are needed as synchrophasor measurement devices are deployed for operation sense-making, decision-making and implementing actionable control.

The optimization and control systems for a smart grid environment will require dynamic information and computational capabilities to handle the uncertainties and variability that exist especially with renewable energy integration. Intelligent technologies needed for sense-making, situational awareness/intelligence, decision-making, control and optimization in a smart grid environment will be presented in this talk.

Prof. Hojjat Adeli (Ohio State University, USA)

“Understanding the Brain: From Automated EEG-Based Diagnosis of the Neurological and Psychiatric Disorders to Brain-Computer Interface”

In this keynote lecture, the author first presents a novel multi-paradigm methodology for automated electroencephalogram (EEG)-based diagnosis of neurological and psychiatric disorders. The methodology is based on adroit integration of three different computing technologies and problem solving paradigms: neural networks, wavelets, and the chaos theory. Examples of the research performed by the author and his associates for automated diagnosis of epilepsy, the Alzheimer’s Disease, Attention Deficit Hyperactivity Disorder (ADHD),  and autism spectrum disorder (ASD) are reviewed briefly. Next, extension of the research into development of a brain-computer interface is discussed. The lecture ends with an outline of research on reading human thought-processes.

TUESDAY

Prof.  Alan Kirman (Aix Marseille University and EHESS, France)

“Is it Rational to have Rational Expectations?”

In economics there has long been a problem within standard theoretical models as to how to treat uncertainty. People are supposed to make their individual decisions based on their perception of the uncertainty with which they are faced. The solution has been to suggest that every individual has the same vision of the process governing the evolution of the economy and that this is consistent with the actual evolution of the economy.

This is manifestly in contradiction with empirical evidence as to peoples’ expectations and with evidence from other disciplines, as to how people form their expectations.

In this presentation I suggest a more simplified and computational approach as to how people learn to modify their expectations and the problems with the convergence of this process. The suggested models are essentially based on computational approaches and derive from Simon’s notions of “bounded rationality”. I will discuss the work so far on this subject and review models with heterogeneous expectations. I will show how the interaction between individuals can influence the formation of collective expectations and give two models which explain market bubbles and crashes using the sort of model I suggest.

Prof.  Ronald Yager (Iona College, USA)

“Social Network Modeling Using Fuzzy Methods”

Web 2.0 has provided for a rapid growth of computer mediated social networks.  Social relational networks provide an important technology for modeling this aspect of human behavioral.  Our goal here is to enrich the domain of social network modeling by introducing ideas from fuzzy sets and related granular computing technologies.  We approach this extension in a number of ways.  One is with the introduction of fuzzy graphs representing the networks.  This allows a generalization of the types of connection between nodes in a network from simply connected or not to weighted or fuzzy connections.  A second and perhaps more interesting extension is the use of the fuzzy set based paradigm of computing with words to provide a bridge between a human network analyst’s linguistic description of social network concepts and the formal model of the network.  We also will describe some methods for sharing information obtained in these types of networks.  In particular we discuss linguistic summarization and tagging methods

Prof. Hani Hagras (University of Essex, England)

“General Type-2 Fuzzy Logic Systems: Towards Higher Order Fuzzy Logic Systems to Handle the Uncertainties in Real World Applications”

Most real world applications face high levels of uncertainties that can affect the operations of such applications. Hence, there is a need to develop different approaches that can handle the available uncertainties and reduce their effects on the given application. To date, Type-1 Fuzzy Logic Systems (FLSs) have been applied with great success to many different real world applications. The traditional type-1 FLS which uses crisp type-1 fuzzy sets cannot handle high levels of uncertainties appropriately. Nevertheless it has been shown that higher order FLSs such as general type-2 FLSs can handle such uncertainties better and thus produce a better performance. However, the immense computational complexities associated with general type-2 FLSs have until recently prevented their application to real world control problems.

This talk will explain the concepts of interval and general type-2 FLSs and will present a new framework to design general type-2 FLS. The proposed approach will lead to a significant reduction in both the complexity and the computational requirements for general type-2 FLSs while offering the capability of representing complex general type-2 fuzzy sets. This talk will explain how the proposed approach can present a way forward for fuzzy systems in real world environments and applications that face high levels of uncertainties. The talk will also present the successful application of type-2 FLSs to many real world settings.

Prof. Yaser Abu-Mostafa (Caltech, USA)

“Machine Learning in a changing environment”

Machine Learning studies how computational systems can automatically learn to perform a desired task based on information extracted from the data. The field is one of the cornerstones of Big Data, and has wide impact on financial, medical, commercial, and scientific applications.

In this talk, I will give an overview of machine learning then focus on one aspect that has become increasingly important with the explosion of recommender systems in e-commerce, as well as financial forecasting applications. This aspect is how can machine learning algorithms perform well when they are trained on data coming from one environment, then tested on data coming from a different environment.

Prof. Boleslaw Szymanski (Rensselaer Polytechnic Institute, USA)

“Influence Spreading and OPinion Dynamics in Social Networks”

Human behavior is profoundly affected by the influenceability of individuals and their social networks. This talk discusses the dynamics of spread of opinions in such networks using two fundamental models for social contagion: the binary agreement model and threshold model. In the first one, all individuals initially adopt either opinion A or B, and some of them commit to their opinion. Committed individuals are immune to influence. We show that the prevailing majority opinion in a population can be rapidly reversed by a small fraction of randomly distributed committed individuals. When committed individuals exist for both opinions, the difference between larger and smaller fractions of them needed for rapid majority conversion decreases as the smaller minority increases. In the threshold mode an initial holders of one opinion try to trigger a cascade of their opinion adoptions. In this context, we seek efficient spreaders and fast heuristic selection strategies, and assess the impact of clustering on system dynamics. We find that  for arbitrarily high threshold, a critical initiator fraction exists beyond which the cascades become global. We show that community structure within the network is more amenable to opinion spread than a homogeneous random network.

WEDNESDAY

Prof. Marco Dorigo (Université Libre de Bruxelles, Belgium)

“Swarm-bots and Swarmanoid”

Swarm robotics is about constructing and controlling swarms of autonomous robots that cooperate to perform tasks that go beyond the capabilities of the single robots in the swarm.  In the talk, I will present the results of two large experiments in swarm robotics: Swarm-bots and Swarmanoid.

In Swarm-bots, I consider a swarm of s-bots, ground robots capable of connecting to, and disconnecting from, other s-bots. When they are connected to each other the s-bots become a single robotic system, a swarm-bot, that can move and change its shape. A swarm-bot can solve problems that cannot be solved by s-bots alone such as transporting heavy objects, moving on rough terrain, and passing obstacles such as holes or steps. I will show video recordings of experiments we performed to study coordinated movement, path formation, self-assembly, collective transport, shape formation, and other collective behaviours using the swarm-bot platform.

In Swarmanoid, I consider a heterogeneous swarm composed of three types of autonomous robots: flying, climbing and ground robots. These robots cooperate both physically and logically: climbing robots cannot move on the ground and are transported to the climbing location by ground robots, while the movements of the ground and climbing robots are guided by the flying robots.  I will present the results of experiments in which the Swarmanoid robots cooperate in a search and retrieval task in a 3-dimensional environment.

Prof. Xin Yao (University of Birmingham, England)

“Cooperative Co-evolution in Tackling Large Optimisation Problems”

Cooperative co-evolution has been used frequently in solving large and complex optimisation problems in recent years. This talk first gives a brief introduction to cooperative co-evolution. Then some recent work is described, including an early attempt at solving large scale optimisation problems,  a cooperative co-evolution particle swarm optimiser (CCPSO), and cooperative co-evolution for combinatorial optimisation, i.e., for the capacitated arc routing problem (CARP). The importance of grouping in cooperative co-evolution will be highlighted. Finally, some future research directions are pointed out.

Prof. José Mendes (Aveiro University, Portugal)

“Structural properties of complex networks”

In this talk I will revisit a number of well-studied problems concerning structural properties of complex networks. Some concepts like percolation, k-core organization, bootstrap percolation are well well-known to the audience but I wilI present them in a different perspective showing the recent analytical advances from a network theory point of view. I will

show that however different problems they share some common features like a hybrid phase transition.

Prof. Russell Eberhart (Indiana University – Purdue University, USA)

“Particle Swarm: From Artificial Life to Cognitive Radio”

Particle swarm optimization has evolved from modeling artificial life and social systems to applications for medicine, security and defense.  Engineering applications for estimating battery state of charge and optimizing container port yard planning were among early successes.  Applications to the fields of extended analog computing, supply chain optimization, and biomedical engineering are more recent examples.  A focus is emerging in the fields of security and defense applications.  Examples of ongoing work are developments in unmanned vehicle mission planning optimization, analysis and diagnosis of human tremor, and resource allocation optimization for cognitive radio.

Prof. Jaime Sichman (University of São Paulo, Brazil)

“A Service Oriented Architecture for Agent Reputation Models Interoperability”

As there is no one single universal accepted model for plenty of agents issues, like organization, reputation, etc., interoperability is a key issue for enabling dealing open MAS, i.e., a MAS where heterogeneous agents may evolve. In this talk, I will present SOARI, an architecture based on ontologies and Web Services to tackle this problem in the reputation domain, We will illustrate the use of SOARI within the ART Testbed.

Prof. José Principe (University of Florida, USA)

“A Cognitive Architecture for Object Recognition in Video”

This talk describes our efforts to abstract from the animal visual system the computational principles to explain images in video. We develop a hierarchical, distributed architecture of dynamical systems that self-organizes to explain the input imagery using an empirical Bayes criterion with sparseness constraints and dual state estimation. The interpretation of the images are mediated through causes that flow top down and change the priors for the bottom up processing. We will present preliminary results and show simplified models for audition.