Log in

Login to your account

Username *
Password *
Remember Me
From Brain Science to Intelligent Machines

Dynamics of recurrent neural network and its implication for higher order cognitive function

Date: Tuesday26/04/2016

Venue: MS020

Time: 10.00 am

Speaker:    Prof. Da-Hui Wang

Affiliation:  Beijing Normal University, China

                 

Dynamics of recurrent neural network and its implication for higher order cognitive function

By 

Prof. Da-Hui Wang

Beijing Normal University, China

Abstract

The cortical part of the brain has plenty of recurrent connectivity, which provides rich repertoire of neural circuit dynamics, and providing neural substrates for higher order cognitive functions such as working memory and decision-making. In this talk, I will discuss how the dynamics of a recurrent network implements multiple-item visual working memory and discrimination between multiple alternative motion directions formed by random moving dots. In the network, neurons are uniformly placed on a ring, labeled by their preferred directions. The connections between excitatory pyramidal cells are structured as a Gaussian function of the difference in the preferred directions, and the connections onto and from the inhibitory interneurons are uniform. The excitatory synaptic currents are mainly mediated by NMDA receptors. This network can simultaneously elicit multiple localized activities after withdrawal of external stimulations, representing the memorization of multiple items. The competition between excitation and inhibition determines the fading/merging of localized activity, furthering the capacity of working memory. The network can be reduced as an analytically tractable, partial integro-differential system. Using nonlinear dynamical analysis, we characterize the multiple-choice decision behaviors and the time course of neural activities underlying decisions, providing a mechanistic explanation for the observations noted in the experiments.

Short Bio

Da-Hui Wang is a Professor in the School of Systems Science and State Key Laboratory for Cognitive Neuroscience and Learning, at Beijing Normal University (BNU), China. He obtained his PhD in systems theory in 2002 from BNU. Dr. Wang focuses on the complexity in neural system, especially, the neural dynamics underlying cognitive function such as working memory, decision-making, and neural modulation.

Using spikes to code sound

Date: Monday 25/04/2016

Venue: MS105 (Boardroom)

Time: 11.00 am

Speaker:    Prof. Leslie Smith

Affiliation:  Stirling University, UK

                 

Using spikes to code sound

By 

Prof. Leslie Smith

Stirling University, UK

Abstract

Much of the sound interpretation world uses Mel-frequency cepstral  coefficients (MFCCs) for coding sound for interpretation. While such coding is highly compressed, it loses all information about the fine time-structure of the sound, and this is particularly important both for locating sound sources and for separating foreground sound from the background. We describe a simple spike (event) based coding technique that maintains fine time structure, and report on how we have used it for sound direction finding, for segmenting sounds, and for determining the type of musical instrument that generated a sound. We discuss why this type of representation can be useful for sound interpretation in realistic environments, and discuss possible ways forward for this type of approach.

Short Bio

Leslie S. Smith: After graduating in Mathematics in 1973, went off to work in the computing industry, but then returned to do a PhD in computing at Glasgow, graduating in 1981. Initially he worked on parallel computing, and taught at Glasgow, before arriving at Stirling University in 1984 and becoming involved in neural networks and computational neuroscience. More recently, he has worked on neuroinformatics, including helping create the CARMEN e-Science portal, and on biologically inspired early auditory processing. Currently, he is Professor of Computing at Stirling University, and, having stepped down two years ago from being Head of Department, has a little more time to devote to research.

Design and control of hand exoskeletons for rehabilitation

Date: Thursday 25/02/2016

Venue: MS105 (Boardroom)

Time: 1.00 pm

Speaker:    Prof. Ashish Dutta
Affiliation:  Department of Mechanical Engineering
                 Indian Institute of Technology Kanpur, India

                 

Design and control of hand exoskeletons for rehabilitation

By 
Prof. Ashish Dutta

Department of Mechanical Engineering
Indian Institute of Technology Kanpur, India

Abstract

Patients suffering from loss of hand functions caused by stroke or spinal cord injuries have driven a surge in the development of wearable assistive devices in recent years. In this presentation the design and control of two finger and three finger exoskeletons are described for rehabilitation of stroke patients. The designs are based on the human finger motion trajectory, in order to emulate the motion of the varying instantaneous axis of rotation of the finger axis. The two finger exoskeletons are capable of simple grasping while the three finger exoskeletons are capable of fine object manipulation, like translation and rotation of small objects. Several techniques have been employed to control the exoskeletons using the subject’s bio-signals like EMG, EEG or inverse kinematics models. Several experimental results with healthy subjects will also be presented.

Short Bio

Prof. Ashish Dutta obtained his PhD in Systems Engineering from Akita University, Japan. From 1994 to 2000 he was with the Bhabha Atomic Research Center (India) where he worked on telemanipulator design and control for nuclear applications. Since 2002 he is with the department of mechanical engineering in the Indian Institute of Technology Kanpur, India. He was also a visiting professor in Nagoya University, Japan in 2006 - 2007and is currently a visiting professor at Kyushu Institute of Technology, Japan. His research interests are in the areas of humanoid robotics, micro sensors and actuators, intelligent control systems and rehabilitation engineering. He has over 100 publications in various international journals, conferences, book chapters, books, etc

Cognition in action and action through cognition: What our motor behaviour and decisions can tell about each other

Date: Wednesday20/04/2016

Venue: MS105 (Boardroom)

Time: 11.00 am

Speaker:    Dr. Arkady Zgonnikov
Affiliation:  School of Psychology, NUI Galway

                 

Cognition in action and action through cognition: What our motor behaviour and decisions can tell about each other

By 

Dr. Arkady Zgonnikov

School of Psychology, NUI Galway

Abstract

This talk will touch on a few links between human decision making and motor behaviour. First, more and more research on decision making shifts from studying static decision outcomes to analysing comprehensive dynamics of decision process. Such research supports the hypothesis that our cognitive processes during decision making can ‘leak’ into motor execution of a decision. Growing amount of experimental data in this field requires new mathematical methods to analyse individual differences in decision dynamics. Second, motor processes traditionally considered as unconscious and automatic (such as balance control during quiet standing) can evoke surprisingly complicated control mechanisms, for instance, intermittent control. A simple motor task, inverted pendulum balancing, provides a number of insights into these mechanisms. In particular, it turns out that the models of switching feedback can benefit from recent research on human cognition; this can have profound implications for real-world problems, e.g. falling in elderly.

Short Bio

Arkady Zgonnikov received MSc in Applied Mathematics from Saint-Petersburg State University, Russia in 2009, and PhD in Computer Science and Engineering from University of Aizu, Japan, in 2014. His PhD research involved experimental and theoretical investigations of human intermittent control. Since 2015, he has worked as a postdoc researcher at School of Psychology in NUI Galway, where he develops mathematical approaches to analysing dynamics of decision making. His current research interests include mathematical modelling, decision making, motor control, and applications of machine learning to human behaviour.

Cognitive Science and Human Factors Research in the Aerospace Domain

Date: Wednesday 10/2/2016

Venue: MS105 (Boardroom)

Time: 11.00 am

Speaker:     Chris Michael Herdman
                   Full Professor of Cognitive Science and Psychology
                   Scientific Director, Centre for Visualization and Simulation (VSIM)
                   Head 
Advanced Cognitive Engineering (ACE) Lab

Affiliation:  Carleton University, Ottawa, Ontario, Canada

                 

Cognitive Science and Human Factors Research in the Aerospace Domain

By 
Prof. Chris Herdman

Full Professor of Cognitive Science and Psychology
                   Scientific Director, Centre for Visualization and Simulation (VSIM) 
                   Head 
Advanced Cognitive Engineering (ACE) Lab

Carleton University, Ottawa, Ontario, Canada

Abstract

TI will present a high-level overview of my lab’s research at the Carleton University Visualization and Simulation (VSIM) Centre.  My intent it to open a dialogue with researchers at the University of Ulster.  The $30M VSIM Centre was established in 2007 through $15M funding from industry and with $15M matching funds from the Canadian government.  The Centre supports multidisciplinary R&D, with strong collaborations between researchers from the human sciences (cognitive science, psychology, HCI) and engineering (aerospace, systems & computer).  The Centre also supports research in immersive media architecture.  New initiatives at the Centre have reached into the humanities, including a project using VR to allow disabled children to create and perform in musical space.  I will summarize a few of the projects that my lab has conducted in the aerospace domain, including assessments of head-up displays (HUDs) and our research on the role of motion cueing systems in flight simulator training of pilots.

 

RESEARCH OVERVIEW

My approach is to link fundamental research on human perception, cognition and attention to the design, implementation and evaluation of advanced human-machine systems. Current areas of research are listed below:   

  • FLIGHT SIMULATOR DESIGN AND PILOT TRAINING

Flight simulators are used extensively to train pilots. The long-term objective of this research line is to develop next-generation flight training systems using virtual reality (VR) technologies.  Fundamental research includes examining the role of motion cueing in simulator-based training experiences as well as issues of embodied cognition associated with using virtual and augmented realty systems for training in complex task domains.

 

  • BRAIN IMAGING AND HUMAN ATTENTION

The long-term objective of this research is to develop an on-line brain monitoring capability that can be used in aircraft to monitor pilot workload and pilot fatigue.  This research is support through a partnership with the Canadian National Research Council (NRC) Flight Research Lab.

  •  AGING PILOTS: COGNITIVE HEALTH AND PILOT PERFORMANCE

The long-term objective of this line of research is to develop and validate a computerized assessment battery that can be used to detect cognitive health issues that may compromise performance and safety with older pilots. Fundamental research is focused on developing a cognition-oriented model of performance for aviation (COMP-A) that specifies the relative impact of pilot attribute factors (age, expertise), domain-independent variables (working memory, processing speed, inhibition, visual-spatial attention), and domain-dependent variables (situation awareness, prospective memory, task management, task-relevant performance) on performance outcomes.

  • CANADIAN SPACE AGENCY

This multi-year project is aimed at developing a game-based cognitive assessment tool to assess the cognitive health and performance readiness of astronauts during long-endurance missions (e.g., International Space Station).  

RESEARCH FUNDING

PRINCIPAL INVESTIGATOR RESEARCH FUNDING - PAST 7 YEARS

$45+ MILLION

Canadian Foundation for Innovation (CFI): VSIM Centre                            $28.8M

Canadian Foundation for Innovation (CFI): HCI Centre                                 $4.6M

Department of National Defense (DND)                                                          $4.5M

Canadian National Search and Rescue Secretariat                                            $2.2M

SSHRC INE Collaborative Grant                                                                     $2.6M

NSERC Funding (Discovery, Strategic, Equipment) (PI)                                 $975K

Canadian Space Agency (CSA)                                                                        $550K

Ontario Centers of Excellence (OCE) (PI)                                                        $745K

GOC Contract for International Assessment of HDTV (PI)                            $250K

 

 

VISUALIZATION AND SIMULATION (VSIM) CENTRE

I am the Scientific Director of the Carleton University VSIM Centre.  As the Principal Investigator, I established the VSIM Centre in 2007 through the acquisition of $30M funding. This included $15M from the Canadian Government and $15M from Industry partners.

Unique Facility and Infrastructure.  The VSIM Centre is an 80,000 square foot special built facility.  The Centre consists of 10 core labs equipped with state-of-the-art visualization, simulation, virtual environment and modeling technologies. Infrastructure includes several flight simulators (helicopter, fixed wing, fast jet), as well as vehicle, naval (ship deck) and UAV simulators. 

Annual Funding.  The VSIM Centre attracts research funding of approximately $5M annually.

 

Multidisciplinary.  The VSIM Centre directly support the research of over 20 faculty members from the Human Sciences (Cognitive Science, Psychology, HCI), Engineering (Systems, Computer, Aerospace), Computer Science, and the School of Architecture (3D immersive media). 

Student Training.  The VSIM Centre supports the training of 80+ graduate students annually.  The students come from seven different academic programs and across three Faculties. The Centre hosts postdoctoral students and numerous visiting faculty and researchers from abroad.  Students often gain practical experience from internships with the VSIM industry partners and with key government agencies.

Industry & Government Partners.  The VSIM has over 25 industry partners. Industry partners are actively engaged in R&D at the Centre.  On average, 20-30 industry employees are onsite at VSIM daily.   The VSIM Centre is directly connected to key government agencies and research labs, including the National Research Council (NRC), Transport Canada, Transportation Safety Board, and Defense R&D Labs.