Call for Participants

SICE System Integration Division, Technical Committee on Ambient Intelligence,
Technical Committee on Interaction and Intelligence


Sponsored by     SICE System Integration Division, Technical Committee on Ambient Intelligence,
                        Technical Committee on Interaction and Intelligence


Date:     Sept. 27, 2010 10:00 ―16:50

Venue:
 
Large Conference Room, International House,
Tokyo Metropolitan University, Minami-Osawa Campus,
http://www.tmu.ac.jp/english/Campus_Map/access.html
http://www.tmu.ac.jp/english/Campus_Map/minami-osawa.html

Participation Fee: Free

Program:


1. 10:00―11:00   
Title: Bacterial Memetic Algorithm and Its Applications
Lecturer: Prof. Botzheim Janos (Szechenyi Istvan University, Hungary)

2. 11:10―12:10   
Title: Biomimetics and Central Pattern Generators (CPGs)
Lecturer: Prof. Maki Habib (The American University in Cairo, Egypt)

3. 13:30―14:30   
Title: 3D Human Motion Analysis via Fuzzy Quantile Inference
Lecturer: Dr. Honghai Liu (University of Portsmouth, UK)

4. 14:40―15:40
Title: Biologically Inspired Methods for Learning in Artificial Systems
Lecturer: Prof. Mattias Wahde (Chalmers University of Technology, Sweden)

5. 15:50―16:50   
Title: Intelligent Human-Centred Vision System
Lecturer: Dr. Chan Chee Seng (University of Malaya, Malaysia)


Contact:
     Naoyuki Kubota           
     Tokyo Metropolitan University
              Organizing Committee Chair
              FAN2010 / iFAN 2010      
     Phone & Fax: +81-42-585-8441   
     E-mail: kubota@tmu.ac.jp

Toru Yamaguchi
Tokyo Metropolitan University
Chair, Technical Committee on Ambient Intelligence
SICE System Integration Division
Phone & Fax: +81-42-585-8644
E-mail: yamachan@tmu.ac.jp

Related Events:
Fan 2010 / iFan 2010, Tokyo metropolitan University, September 25-26, 2010
http://www.sd.tmu.ac.jp/ifan2010/

PDF Files:
        Access
        Abstract
        All

Access to Tokyo Metropolitan University:
http://www.tmu.ac.jp/english/Campus_Map/access.html


Address: 1-1 Minami-Osawa, Hachioji-shi, Tokyo, Japan 192-0397
Tel: +81-42-677-1111
Access: 5-minute walk from the ticket gate at the Minami Osawa Station, Keio Sagamihara Line.



http://www.tmu.ac.jp/english/Campus_Map/minami-osawa.html







19. International House



1. 10:00―11:00    Title: Bacterial memetic algorithm and its applications
Lecturer:  Prof. Botzheim Janos (Szechenyi Istvan University, Hungary)


Abstract:  Evolutionary algorithms play a significant role as search techniques for handling complex problems in many fields such as artificial intelligence and engineering. The advantage of evolutionary algorithms is their ability to solve and quasi-optimize problems with non-linear, high-dimensional, multi-modal, and discontinuous character. These algorithms have the ability to explore large spaces, without demanding the use of derivatives of the objective functions, such as by gradient-based training methods. Their principles are based on the search for a population of solutions, where tuning is done using mechanisms similar to biological recombination. Over the last decade, there has been increasingly interest in applying bacterial evolutionary algorithms [1] for solving optimization problems. The bacterial evolutionary algorithm incorporates two operators based on microbial evolution phenomenon. The bacterial mutation optimizes the bacteria individually, whilst the gene transfer allows the bacteria to directly transfer information to other bacteria in the population. Evolutionary techniques explore the whole objective function, because of their characteristic, so they find the global optimum, but they approach to it slowly. Memetic algorithms [2] combine an evolutionary algorithm with a local search procedure in order to speed up the evolutionary process making it more efficient and find a better, more accurate solution. In this presentation these ideas are combined and the bacterial memetic algorithm [3] is introduced. This technique applies the bacterial approach incorporating a local search procedure. The bacterial memetic algorithm can be applied for the optimization of fuzzy rule bases. In this task the most effective local search methods are based on gradient calculation. One of the best gradient-based technique, the Levenberg-Marquardt algorithm is suggested as local search procedure within the bacterial memetic algorithm. Another application of the bacterial memetic algorithm is the approximate solution of the Traveling Salesman Problem and its modified version in which the requirements and features of practical application in road transportation and supply chains are taken into consideration [4, 5].



2. 11:10―12:10    Title: Biomimetics and Central Pattern Generators (CPGs)
Lecturer: Prof. Maki Habib (The American University in Cairo, Egypt)


Abstract:  This talk presents the background of biomimetics approach that inspires the development of new technologies and techniques. Then, it introduces the design and analysis of a controller based on a biologically inspired central pattern generator (CPG) network of mutually coupled Matsuoka nonlinear oscillators to generate adaptive rhythmic human like movement for biped robots. The talk focuses on the way in which the sensory signals feedback contribute to generate dynamic, stable and sustained rhythmic movements with robust gaits for biped robots. In addition, the talk shows how the driving input and external perturbation affect the speed of locomotion and change the period of its own active phase. The new design was studied through interaction between simulated interconnection coupling dynamics with 6 links and a musculoskeletal model with 6 degrees of freedom (DOFs) of a biped robot.. The simulated model helps to realize the interaction between the controller, the mechanism of the robot, and the environment.  In addition, it helps to study the necessary conditions for efficient generation of stable rhythmic walking at different speed, on different type of terrains and robustness in response to disturbances. Evaluations of the developed CPG based adaptive bipedal locomotion are carried out through simulations with successful testing results.
 
3. 13:30―14:30    Title: 3D Human Motion Analysis via Fuzzy Quantile Inference
Lecturer: Dr. Honghai Liu (University of Portsmouth, UK)


Abstract:  Enormous uncertainties in unconstrained human motions lead to a fundamental challenge that many recognising algorithms have to face in practice: efficient and correct motion recognition is a demanding task, especially when human kinematic motions are subject to variations of execution in the spatial and temporal domains, heavily overlap with each other, and are occluded. Due to the lack of a good solution to these problems, many existing methods tend to be either effective but computationally intensive or efficient but vulnerable to misclassification.
This talk presents a novel inference engine for recognising occluded 3D human motion assisted by the recognition context. First, uncertainties are wrapped into a fuzzy membership function via a novel method called Fuzzy Quantile Generation which employs metrics derived from the probabilistic quantile function. Then, time-dependent and context-aware rules are produced via a genetic programming to smooth the qualitative outputs represented by fuzzy membership functions. Finally, occlusion in motion recognition is taken care of by introducing new procedures for feature selection and feature reconstruction. Experimental results demonstrate the effectiveness of the proposed framework on motion capture data from real boxers in terms of fuzzy membership generation, context-aware rule generation, and motion occlusion. Future work might involve the extension of Fuzzy Quantile Generation in order to automate the choice of a probability distribution, the enhancement of temporal pattern recognition with probabilistic paradigms, the optimization of the occlusion module, and the adaptation of the present framework to different application domains.



4. 14:40―15:40    Title: Biologically inspired methods for learning in artificial systems
Lecturer: Prof. Mattias Wahde (Chalmers University of Technology, Sweden)


Abstract:  Learning, i.e. the ability to process sensory input, and to store relevant information for future use, is one of the most important factors in adaptive behavior, both in biological and artificial organisms.
Much of what is known about learning on a neural level has been obtained from biological studies of model organisms such as the sea slug Aplysia and the worm C. Elegans. These studies, in turn, have inspired researchers to develop learning methods for artificial systems, using similar principles as those found in biological systems.
In this talk, I will start by reviewing basic learning in biological organisms, and I will then present some computational models of basic learning and their applications.


5. 15:50―16:50    Title: Intelligent Human-Centred Vision System
Lecturer: Dr Chan Chee Seng (University of Malaya, Malaysia)


Abstract:  Human learning is clearly not limited to any single strategy, but can involve any type of strategy, or a combination of them, depending on the task at hand. Research on multi-strategy learning is therefore a key to understanding learning processes in general, to making progress in machine learning, as well as to extending the applicability of current machine learning methods to new practical domains. This will motivate the use of multiple sensors in the environment as well as contextual information for effective data and decision fusion. This talk will focus on the user interaction techniques formulated from the perspective of key human factors such as adaptation to user preferences and behavior models.


The list of lectures and seminars