|
Seminar
Schedule for Fall 2004 |
||||
Date |
Speaker |
Title |
Institution |
Location |
9/24/04 |
Morgan McGuire |
Computer graphics
and game design |
Brown University |
DION 101 |
10/8/04 |
Li Shen | Spherical
Parameterization for 3D Closed Surfaces and Its applications |
UMassD |
DION 101 |
10/22/04 |
Boleslaw Mikolajczak | Verification and
Validation of Concurrent Object-Oriented Systems using Colored Petri
Nets |
UMassD |
DION 101 |
11/5/04 |
Vinod Vokkarane |
Multi-Layer
Survivability in IP-over-Optical Networks; Quality of Service in
Optical Burst-Switched Networks" |
UMassD |
DION 101 |
11/12/04 |
Shelley Zhang |
The formation of virtual organization | UMassD |
DION 101 |
11/19/04 |
Ramprasad
Bala |
Mesoscale
Feature Tracking and their role in Climate Studies |
UMassD |
DION 101 |
12/3/04 |
Graduate
Students |
Research
Assistant presentation on research projects |
UMassD |
DION 101 |
|
Seminar
Schedule for Spring 2004 |
||||
Date |
Speaker |
Title |
Institution |
Location |
3/5/04 |
Jurgen Schulze |
Volume rendering
in the CAVE |
Brown University |
DION 101 |
3/26/04 |
Sol Neeman | Arts, Computers and
Artificial intelligence |
Johnson and Wales University |
DION 101 |
4/9/04 |
Haiping Xu | A Security Based Model
for Mobile Agent Software Systems |
UMassD |
DION 101 |
4/23/04 |
Emad Aboelela |
TCP: From Static to
Mobile Networks |
UMassD |
DION 101 |
5/7/04 |
Jan Bergandy |
Design Patterns and Software Development Process |
UMassD |
DION 101 |
In the talk I will present my research about interactive visualization of volumetric datasets in virtual environments. The three major topics will be:volume rendering algorithms, rendering with parallel computers, and a user interface for the CAVE. Direct volume rendering is the display of three dimensional scalar fields without prior data conversion to polygonal models. For example, a computer tomography (CT) scan from a hospital consists of a number of images, which the computer can stack on top of each other to reconstruct the original 3D object. This display method requires a great deal of computing power because for each reconstructed 3D image every pixel of each 2D image needs to be taken into account. Parallel computers can be used to increase the rendering speed of large volumetric datasets. I will present a new method that runs the CPU-based shear-warp algorithm on a parallel computer to show the results in a CAVE. The CAVE is a small room whose walls are projection screens. Typically, four screens are projected on: the front, left, and right walls, and the floor. Stereo rendering allows for virtual reality applications. The users interact with a 3D mouse which transmits its location and orientation to the computer.
Seminar
Schedule for Fall 2003 |
||||
Date |
Speaker |
Title |
Institution |
Location |
9/26/03 |
Ram Bala |
Visualization and Physical
Oceanography |
UMassD |
DION 101 |
10/10/03 |
Iren Valova |
Spiking Neurons for
Modeling Olfactory Bulb Functionality |
UMassD |
DION 101 |
10/24/03 |
Don Towsley |
Internet worms: modeling
and detection |
UMass, Amherst |
DION 101 |
11/21/03 |
Adam Hausknecht |
Developement of Math
Software |
UMassD |
DION 101 |
12/05/03 |
Bolek Mikolajczak |
Study of Structural and
Behavioral Petri net Properties with respect to Abstraction and
Refinement |
UMassD |
DION 101 |
Date: December 5, 2003 download
Speaker: Boleslaw Mikolajczak
Title: Study of Structural and Behavioral Petri net Properties with respect to Operations of Abstraction and Refinement and their Role in Development of Concurrent Systems
Abstract:We discuss a role of Petri net morphisms in conceptual modeling of concurrentDate: September 26, 2003
computing systems with several levels of abstractions and/or refinements. We study
structural and behavioral properties of concurrent systems that are modeled by Petri nets.
These properties, particularly liveness, boundedness and safety, are important in
assuring correctness of design for complex concurrent systems. Examples
are also provided to illustrate significance and applications of Petri net morphisms
in development of concurrent systems.
Speaker: Dr. Ramprasad Bala
Title: Visualization, Detection and Tracking of Mesoscale Oceanic Features in an Ocean Circulation Model
Abstract:The process of identifying and tracking oceanic eddies over space and time, and their relationship to the net poleward heat transport are of fundamental importance for Climate studies. The visualization of heat flux in the numeric motivates the detection and tracking of mesoscale features to understand their role in poleward heat transport. The numeric model MICOM for the North Atlantic is used. Methods developed to automatically detect, locate and track mesoscale eddies spatially and temporally are presented. Using structuring elements based on velocity information eddy centers are detected and streamline plots used to eliminate false positives. Additional visualization, based on the detected eddy centers provides information about eddies temporally and spatially.Date: April 18,2003 - download
Speaker: Boleslaw Mikolajczak
Title: Integrating Object-Oriented Design and High-level Petri Nets in Software Development of Concurrent Systems
Abstract:A method of designing concurrent systems, based on integration of Object-Oriented Design and High-level Petri nets, is presented. Object-Oriented technology has well established design techniques but lacks analysis, verification and validation methods of the designed system and formal specification of concurrency and conflict. All these features are extremely important from practical point of view, especially in terms of rapid prototyping of parallel and distributed systems. On the other hand, Petri net is a well defined formalism of parallel and distributed systems wrt modeling, verification, validation (including conflict resolution and resource allocation) that lacks clear and effective design techniques. The method takes advantages of both Object-Oriented Design and High-level Petri nets by using Object-Oriented methodology on the design stage and High-level Petri nets on the analysis, verification and validation stages. Petri net modeling of classes with inheritance and solution to inheritance anomaly are central to achieve executable specification of the OO system. The method utilizes specialized CASE software tools that effectively support modeling, verification, and validation of high-level Petri nets, such as Design/CPN and SYROCO.
Date: April 04,2003 - download
Speaker: Xiaqin Shelley Zhang
Title: Integrative Negotiation In Complex Organizational Agent Systems
Abstract:Agents are autonomous, intelligent, computational entities that pursue goals or carry out tasks in order to meet their designed objectives. A multi-agent system (MAS) consists of a group of agents that can take specific roles within organizational structures. Computers are no longer stand-alone devices; they are tightly connected to each other and their users. Researchers have applied the technology of autonomous agents and multi-agent systems in many applications, such as airport management, traffic control and transport logistics, and distributed electricity management. These applications require a number of agents to work together, given the complexity of these problems, also the distribution of the data and resources. Organization structures are necessary for large-scale agent-based systems to operate efficiently. Meanwhile, organizational structures in a large-scale agent society are not static. Agents from different, separate organizational entities will come together to dynamically form virtual organizations for solving specific problems that are relevant to each of their organizational entities. This presentation addresses the problem of negotiation in a complex organizational context and tries to bridge the gap between self-interested negotiation and cooperative negotiation. An integrative negotiation mechanism is introduced, which enables agents to choose any attitude from the extremes of self-interested and cooperative to those that are partially self-interested and partially cooperative. This mechanism is based on and also extends the motivational qualities (MQ) framework for evaluating which task an agent should pursue at each time point. Experimental work verifies this mechanism and explores the question whether it always improves the social welfare to have an agent be completely cooperative.
Date: March 28,2003 - download
Speaker:Emad Aboelela
Title: Quality-of-Service Provision For Multimedia Networking
Abstract:For many years, packet-switched networks have offered the promise of supporting multimedia applications, that is, those that combine audio, video, and data. Audio and video applications are examples of real-time applications. The best-effort model, in which the network tries to deliver your data but makes no promises and leaves the "cleanup operation" to the edges, is not sufficient for real-time applications. What we need is a new service model; one in which applications that need better assurances can request such service from the network. The network may then respond by providing an assurance that it will do better or perhaps by saying that it cannot promise anything better at the moment. A network that can provide different levels of service is often said to support Quality of Service (QoS). In this seminar different developed approaches to provide a range of QoS will be illustrated. An optimal routing model will be presented. The routing model is formulated as a fuzzy optimization model, that takes into consideration bandwidth allocation to avoid link saturation and hence the possibility of congestion. A management-dependent margin is used for bandwidth allocation. A path is considered attractive to incoming calls as long as its available bandwidth is around that margin. This greatly reduces the fragmentation problem of the load-balancing techniques and avoids packing some paths with heavy loads (with the benefit of decreasing the likelihood of congestion)
Date:November 22, 2002 -
Speaker:Iren Valova
Title:A Growing Parallel Self-Organizing Map for Unsupervised Learning
Abstract:Self-Organizing maps (SOM) have become quite popular for tasks in data visualization, pattern classification or natural language processing and can be seen as one of the major concepts for artificial neural networks of today. Their general idea is it to approximate a high dimensional - previously unknown - input distribution by a lower dimensional neural network structure with the goal to model the topology of the input space as good as possible. Classical Self-Organizing Maps read the input values in random but sequential order one by one and thus adjust the network structure over space: The network will be build while reading bigger and bigger parts of the input. In contrast to this approach, we present a Self-Organizing Map that processes the whole input in parallel and organizes itself over time. The main reason for parallel input processing lies in the fact that knowledge can be used to recognize parts of patterns in the input space that have already been learned. This way, networks can be developed that don't reorganize their structure from scratch every time a new set of input vectors is presented but rather adjust their internal architecture in accordance with previous mappings. One basic application could be a modeling of the whole-part relationship through layered architectures.