Friday, June 3, 2011
1:00 PM - 2:00 PM (Poster Sessions) in UW1-010/020
2:15 PM - 4:30 PM (Orial Presentations) in UW1-010/020/051
One of our largest CSS 497 Colloquiums since 2005, this year’s event will be full of exciting individual and industry sponsored projects that our CSS students have been working on for the past couple of quarters.
As an added bonus, three Applied Computing students will also be presenting their research projects at this event.
The event will be broken down into two segments. The first will be a poster session where the student will be available to talk about and answer questions with regards to their project. The second segment will be a formal 10-15 minute oral presentation given by the student in which he/she will further discuss their project.
A detailed schedule can be found here.
Thursday, May 26, 2011
Social: 5:00pm - 5:30pm (UW1-370)
Presentation: 5:45pm - 6:45pm (UW2-005)
Director of Engineering Agility & UX
Companies worldwide are adopting agile and continuous improvement methods—such as Scrum, A3, Kanban, Theory of Constraints—to organize software development. Under traditional project management, Citrix Online saw project time increase from 6 months in 2003 to 18 months in 2007. After adopting agile methods in 2008, project time has dropped to 7 months and continues to decrease. Agile methods radically change job responsibilities and increase internal transparency; resulting in middle level management resistance. Conscious culture-building was required at Citrix Online to successfully adopt agile methods.
This talk will summarize the key principles of agile methods for teams of 3 to 7 people, and then show how the principles can be scaled fractally for larger organizations. Agile methods drive changes in software engineering practices, product management, executive behavior, facilities design and hiring.
About Dan Greening
Dan Greening holds a PhD from UCLA Computer Science Department, specializing in highly-parallel systems. He studied parallel simulated annealing at IBM Research. He spent many years as a serial entrepreneur, co-founding two successful companies and one dud. He created the first Scrum team at Citrix Online, and is now its Director of Engineering Agility and User Experience.
Computer Aided Software Testing, and Biomimicry: Ways to More Effectively Create Software-Enabled Systems
Thursday, May 19, 2011
8:00 PM - 8:30 PM
Dr. David Socha
My interests, research and teaching is on how to create effective software enabled systems that solve important human needs. This is a holistic design enterprise involving many aspects from technical to social, business to science, quantitative to qualitative. This talk will illustrate two areas that I am focusing on for my research: software testing, and biomimicry. In software testing, I am working with industrial colleagues to explore various aspects of what we are now calling computer aided software testing. This is a simple and effective way of combining the best of what humans do well and what computers do well. One question we are exploring is why some people embrace these techniques, while others do not. In biomimicry, I am exploring how the biomimicry design process is, and could be, used to create better software systems and better organizational processes. There are a variety of ways in which students could be involved in these inquiries, which we can explore during and after the talk.
Tuesday, May 17, 2011
Social: 3:00pm - 3:30pm (UW1-370)
Presentation: 3:30pm - 4:30pm (CC1-041)
Director of the USC GamePipe Laboratory
ACM Distinguished Speaker
The USC Department of Computer Science is in its third year of operating its BS in Computer Science (Games) and MS in Computer Science (Game Development) degree programs. We have developed an interesting educational architecture inside of that degree program that allows the students to become strong game developers, strong computer scientists, strong programmers, strong systems developers, and facile with working in cross-disciplinary, collaborative groups. We believe that educating students in this fashion strengthens our department’s ability to do cutting edge research in computer science as well as provide great graduates for the game industry. In this talk, we share our lessons learned, some detail on our courses and processes, as well as current research and development directions. We discuss and frame an R&D agenda that lead towards interesting potentials for the future of games – massively multiplayer online games (MMOGs) as social model testbeds, sensor-based games, human-aware computing, and emotion-cognizant games.
About Michael Zyda
Professor Zyda began his career in Computer Graphics in 1973 as part of an undergraduate research group, the Senses Bureau, at the University of California, San Diego. Professor Zyda received a BA in Bioengineering from the University of California, San Diego in La Jolla in 1976, an MS in Computer Science from the University of Massachusetts, Amherst in 1978 and a DSc in Computer Science from Washington University, St. Louis, Missouri in 1984.
Professor Zyda has consulted for the White House Office of Science and Technology Policy, the Ministry of Industrial Development Sabah Province, Malaysia, Japan Tech Services Corporation, Tokyo, Hitachi Plant Construction & Engineering, Ohtsuka, SimGraphics Engineering, Pasadena, BBN, Silicon Graphics International, Geneva, Nihon Silicon Graphics KK, Advanced Telecommunications Inc., TecMagik, Muse3d.com, Time Warner, Paramount Digital Entertainment, BBN, MaK Technologies, Sony Computer Entertainment (through Shook, Hardy & Bacon), among others. He is a speaker with Celebrity Speakers, International. He is an advisor to Emsense, Big Stage, iSportGames, Virtual Heroes, and Dragonfire Productions. He is the founder of Westside Transmedia, a company focused on translating film story to the game medium. Learn more...
A Platform for Parallel Processing of Machine Learning Problems in the Amazon Cloud
Wednesday, May 11, 2011
Social: 5:00pm - 5:30pm (UW1-360)
Presentation: 5:45pm - 6:45pm (UW1-320)
Machine learning techniques have become very powerful, and are in widespread use for predicting outcomes in domains such as medicine, industrial processes, scientific research, and finance. The training of machine learning models can be a computational challenge: training sets may contain millions or even billions of records, or, alternatively, the best model may consist of an ensemble of base models, in which the available data is reused hundreds or thousands of times. In either situation, the training is usually amenable to parallel processing.
In this talk I will describe Ensemble Cloud Army (ECA), an open-source platform which supports parallel processing of machine learning problems in the Amazon Cloud. ECA is controlled by a Python script running on the user's local computer, which automatically performs the following steps: 1) reads a local configuration file, 2) instantiates a cluster of EC2 nodes in the Amazon Cloud, 3) establishes intercommunication among the nodes, 4) uploads the user's data files and machine learning code to the cloud, 5) executes the machine learning code in parallel in the cloud, and 6) retrieves the results to the local machine. The first generation of ECA, just released on SourceForge, is designed specifically for ensemble machine learning calculations, and uses MPI (message passing interface) for internode communication and the R statistical language for machine learning computations. A future version under development will substitute Amazon's Elastic MapReduce for MPI.
I will also briefly describe other ways in which our company is using the Cloud Army framework for parallel computations on very large biomedical datasets. ECA was developed under a SBIR (Small Business Innovation Research) grant from the National Institutes of Health.
About James Howbert
Jeff obtained a Ph.D. in synthetic organic chemistry from Harvard University in 1983. He spent the next 25 years in the pharmaceutical and biotech industries, conducting medicinal chemistry and drug discovery research in multiple therapeutic areas. He has been Vice-President of Research at two Seattle-area biotech companies (Celltech and CEPTYR), and is responsible for advancing 6 compounds into clinical development as treatments for cancer, psychiatric diseases, pain, diabetes, and allergy. Jeff recently decided to pursue a second career in computational biology, and completed a M.S. in computer science at University of Washington in 2008. Since then he has worked at two local companies (Insilicos and NeuroVista), building disease prediction models from complex biomedical datasets. He has a particular love of machine learning, and was a member of the team that placed second in the Netflix Prize contest.
Tuesday, April 26, 2011
Social: 5:00pm - 5:30pm (UW1-370)
Presentation: 5:45pm - 6:45pm (UW1-041)
Co-founder & CTO
Most organizations today are dealing with an exponential increase in the amount of data that is collected or stored during routine operations. Complicating this further is the fact that this data may be in a variety of heterogeneous databases such as in custom databases or in enterprise applications -- enterprise resource planning (ERP) or customer relationship management (CRM). This large amount of heterogeneous data leads to some common challenges that users at most organizations face:
- Operational errors and delays due to lack of a single version of the truth for the data.
- Lost productivity from users accessing multiple systems.
- Poor business decisions made from incorrect or incomplete data.
Data integration deals with the effort of combining data residing in the different data sources so that it can be consumed by either one of the systems or other client applications. The difficulty in data integration comes from the fact that each data source has its own schema, semantic definitions, and logical rules. There are multiple approaches for data integration depending on how tightly or how loosely you want the data coupled and what the scope of application that consumes the combined data is. A key aspect of data integration may be the extract, transform, and load (ETL) process and a key technology enabling the data integration process is the service-oriented architecture (SOA).
This talk will cover the challenges of data integration and will introduce the methodology and some of the tools used to address these challenges today. Specific examples from the speaker's experience with SAP data integration will be presented along with some live-demonstrations.
About Vikram Chalana
Vikram Chalana is Co-founder and Chief Technology Officer of Winshuttle. As CTO, he is ultimately responsible for spearheading the current and future direction of the company's technology offerings.
Vikram has spent several years in the software industry concentrating on data mining and business intelligence applications. His primary focus the last few years has been on applications that enable SAP users to greatly improve their ability to integrate and migrate data. His previous position before co-founding Winshuttle was as Director of R & D at Ultrasound, prior to which, he held senior development positions with Insightful Corporation.
Vikram is a published author of more than 20 papers and presentations, and holds more than a dozen US and international patents.
Vikram earned a Bachelor's degree in electrical engineering from the Indian Institute of Technology, Kanpur, India and a Doctorate of Philosophy in Bioengineering with Minors in Electrical Engineering, Computer Science and Statistics from the University of Washington, Seattle, WA.
Sensor Grid Integration: An Agent-Based Workbench for On-the-Fly Sensor-Data Analysis
Wednesday, April 20, 2011
8:00 PM - 8:30 PM
Dr. Munehiro Fukuda
This research project focuses on integrating wireless sensor networks into grid and cloud computing systems, and gives an agent-based workbench for facilitating on-the-fly sensor-data analysis on top of these computing systems.
Although the emergent popularity of cloud computing has allowed users in SME (small and medium-sized enterprise) and SOHO (small office/home office) to construct their business-specific computing systems with cloud services, the cloud-computing users still encounter some practical difficulties in particular when handling a large amount of sensor data: remote job execution, on-the-fly data analysis with parallelization, and sensor-data delivery.
This research targets frost protection as an example application where tree-fruit growers use a temperature sensor network so as to observe and to possibly predict every overnight transition of their orchard temperatures. While they can use various cloud services including remote storages, computing power, and even programming tools, they still need to address by themselves the following three problems: (1) automating the execution of a temperature-prediction program upon arrivals of new temperature data, (2) accelerating the program execution using cloud-offered programming tools, and (3) forwarding sensor data to the program and saving outputs into file servers. Without proper solutions, tree-fruit growers would end up with repeatedly checking the current orchard temperatures.
For more information, please visit the Distributed Systems Laboratory.
Managing Information in Software Engineering and e-Science
Thursday, April 14, 2011
8:00 PM - 8:30 PM
Dr. Hazeline Asuncion
Managing related information is a fundamental task in many contexts. In software engineering, relating the design to requirements is necessary in ensuring that the system to be developed meets customer requirements. In e-Science, identifying the relationships between intermediate data sets is necessary in the repeatability of experiments. Previous work on traceability provided means for tracing across distributed and heterogeneous artifacts across the software development life cycle. Current research investigations entail (1) building traceability support to facilitate change management in software development and (2) applying traceability techniques to the domain of e-Science. This talk will include possible projects within these two threads of research.
Computer Vision at UW Bothell
Thursday, March 31, 2011
8:00 PM - 8:30 PM
Dr. Clark Olson
Dr. Olson will be giving a brief overview of computer vision and describe some of his recent projects, including terrain mapping for Mars exploration, registration of aerial images, and projective clustering algorithms.
Monday, March 7, 2011
5:00pm - 5:45pm (Social Hour @ UW1-370)
5:45pm - 6:45pm (Presentation @ UW1-220)
Senior User Experience Lead
Windows Server (Microsoft)
Although Computer department in universities have been around for decades, a Usability (UX) or HCI Major was non-existent during my college years. With many programs in existence now, it is easy to realize the long way the UX field has come in terms of growing the discipline and injecting plurality and specialization. Users have changed too, and so have the devices and the experiences that people are now accustomed to. The goal of this talk is to help students recognize the ever changing landscape of the UX discipline and to look at it from several perspectives. I will also discuss current trends and suggest how students might position themselves to take advantage of these opportunities.
About Frédéric François
Frédéric François is a Senior User Experience Lead in charge of Windows Server at Microsoft. He has been working for 20 years in the design of user experiences. His career spanned a wide range of companies, from atomic startups to giant corporations, and domains like industrial automation, IT management and medical applications, with one thing in common: the representation and interaction of complex information systems. He holds a Diplôme d’Ingénieur d’Etat from ESIEE (Ecole Supérieure d’Ingénieurs en Electronique et Electrotechnique) Paris, France, with majors in Production Systems Engineering and Computer Science.
GPU Computing for Neural Simulation
Thursday, March 3, 2011
8:00 PM - 8:30 PM
Dr. Michael Stiber
How do brain cells grow the interconnections that allow them to perform useful functions for organisms? When neuroscientists create cultures of neurons and allow them to develop networks in artificial settings, they see networks of cells that produce pathological behaviors that seemingly prevent them from doing anything useful.
The UW Bothell Biocomputing Laboratory (BCL) has been developing simulations of this development process to help understand why these experiments don't work like real brains and what can be done to fix this. The enabling technology for these experiments is GPU (Graphics Processing Unit) hardware, which harnesses the number crunching hardware behind video games and animated movies for general-purpose computation. Prof. Stiber will describe student research opportunities in neural modeling, simulation, data analysis, parallel algorithms, and general-purpose GPU software development.
This event is co-hosted by UW Bothell's Computing & Software Systems (CSS) and Science & Technology (S&T) departments.
Friday, February 18, 2011
Dr. Charles Ofria
Michigan State University
When Darwin first proposed his theory of evolution by natural selection, he realized that it had a problem explaining the origins of traits of "extreme perfection and complication" such as the vertebrate eye. Critics of Darwin's theory have latched onto this perceived flaw as a proof that Darwinian evolution is impossible. In anticipation of this issue, Darwin described the perfect data needed to understand this process, but lamented that such data are "scarcely ever possible" to obtain. In this talk, I will discuss research where we use digital organisms (populations of self-replicating and evolving computer programs) to elucidate the genetic and evolutionary processes by which new, highly-complex traits arise, drawing inspiration directly from Darwin's wistful thinking and hypotheses. I will also explore some of the implications of this research to other aspects of evolutionary biology and new ways that these evolutionary principles can be applied toward solving engineering problems.
About Dr. Ofria
Dr. Charles Ofria is an associate professor at Michigan State University in the Computer Science Department and the Ecology, Evolutionary Biology, and Behavior Program. He has a multidisciplinary background, receiving a PhD from the Computation and Neural Systems department at Caltech, followed by a three-year postdoc in the Center for Microbial Ecology at MSU with evolutionary biologist Richard Lenski. He is currently the director of the MSU Digital Evolution Laboratory and the deputy director of the BEACON Center for the Study of Evolution in Action, a new multi-university NSF Science and Technology Center. His research uses digital organisms to answer fundamental questions in evolutionary biology and harness the results to solve more applied, computational problems.
See http://devolab.cse.msu.edu/ for more information.
Wednesday, February 16, 2011
8:00 PM - 8:30 PM
Dr. Kelvin Sung
Kelvin will talk about two threads of his research directions: serious games, and ultra reality.
Serious games is the study of how to design/build applications that integrate the engaging nature of videogames to accomplish real world objectives, objectives that are beyond the traditional “for fun” entertainment purpose. Working with the UW Bothell Office of Admissions and the Center for Serious Play, Kelvin and his students have built a series of games to assist potential visitors to tour UW Bothell campus and to assist new students on their first day on campus. A few current efforts related to this work will be highlighted, including: the tools required for the automation of building serious games, software development process for designing/building serious games, and current/future serious games that will be build.
Ultra reality is the study of how to reconstruct reality remotely based on samples of the real world. In very limited ways, we can already achieve this goal, e.g., we can view videos from a web-cam and observe reality remotely. The objective of our study is to generalize this paradigm by reconstructing the entire 3D world. For example, imagine a webpage showing you a webcam view of a classroom. Now imagine from home you can manipulate and move the webcam to any new positions around the entire classroom with any new orientations and observing what is happening in the classroom right at that moment, life, in real time. Of course, our study is how to accomplish these manipulations without moving any actual cameras. An overview of our current efforts will be presented.
Wednesday, February 9, 2011
5:00pm - 5:45pm in UW1-361 (Social hour)
5:45pm - 6:45pm in UW1-020 (Presentation)
Consultant & Entrepreneur
Building a 100 MPG car in 3 months with $100K -- Managing a collaborative multi-national team in real-time using Agile/Lean/Scrum/Xp.
Joe Justice, a Seattle-area lean-software consultant and entrepreneur, shares how he ported software-team best practices back to their roots to compete for $10 million in the Progressive Insurance Automotive X Prize. Driven by a desire to optimize automotive performance while minimizing cost and environmental impact, Joe formed WIKISPEED, a small, volunteer-driven team. They are manufacturing a revolutionary 100 mpg, gasoline powered, four-seat car with a target price of $17,995. Joe will walk through how they are accomplishing the seemingly impossible. Joe will explain Agile applied by using his experience in the Progressive Insurance Automotive X Prize as the example.
About Mr. Justice
Joe Justice is a Seattle-area lean-software consultant and entrepreneur, and a registered automotive manufacture since 2007. In 2010, Joe's X Prize team, WIKISPEED, tied for 10th place in the mainstream class of the Progressive Insurance Automotive X Prize, a $10 million challenge for 100+ MPGe automobiles. Joe has spoken on social web application development, project methodology, and agile best practices to audiences at Denver University, University of California Berkley, Google, The Bill and Melinda Gates Foundation, Rotary International, and others. Joe is currently on assignment at Microsoft and CEO of WIKISPEED.
Tuesday, January 25, 2011
5:45pm - 6:45pm
Director of Education & Scholarly Communication
Microsoft External Research
Microsoft External Research strongly supports the process of research and its role in the innovation ecosystem. We are dedicated to the development and support of efforts in open access, open tools, open technology, and interoperability. We partner with universities, national libraries, publishers, and governmental organizations to help develop software and services in order to evolve the process of scholarly information capture and dissemination. Such collaborations demonstrate our ongoing work towards producing next-generation documents that increase productivity and empower authors to increase the discoverability and appropriate re-use of their work. Throughout the research process, software can and should assist us in the process of information capture, organization, analysis, collaboration, authoring, dissemination and long-term information preservation. This session will highlight several freely available and open source efforts from Microsoft External Research, and will demonstrate how these can help to enhance the process of scholarly communication.
About Mr. Dirks
Lee Dirks is the Director of Education & Scholarly Communication in Microsoft’s External Research division, where he manages a variety of research programs related to open access to research data, interoperability of archives and repositories, preservation of digital information as well as the application of new technologies to facilitate teaching and learning in higher education.
A 20+ year veteran across multiple information management fields, Lee holds an M.L.S. degree from the University of North Carolina-Chapel Hill as well as a post-masters degree in Preservation Administration from Columbia University. In addition to past positions at Columbia and with OCLC (Preservation Resources), Lee has held a variety of roles at Microsoft since joining the company in 1996 - namely as the corporate archivist, then corporate librarian, and as a senior manager in the corporate market research organization. During his career, his team's work on the Microsoft Corporate Library intranet site at Microsoft was recognized as a "Center of Excellence Award for Technology" in 2003 by the Special Library Association's (SLA) Business & Finance Division. Additionally, Lee was presented the 2006 Microsoft Marketing Excellence Award by Microsoft CEO Steve Ballmer – for a marketing & engineering partnership around a breakthrough market opportunity analysis process which is now a standard operating procedure across Microsoft.
In addition to participation on several (US) National Science Foundation task forces, Lee also teaches as adjunct faculty at the iSchool at the University of Washington, and serves on the advisory boards for the University of Washington Libraries, the UW iSchool's Master of Science in Information Science (MSIM) program and the Metadata Research Center (MRC) at the School of Information and Library Science at UNC at Chapel Hill.
The Power of Networking
Monday January 24, 2011
5:00 PM - 6:00 PM
President and CEO
Workforce Development Council Snohomish County
Knowledge and relationship based mentoring has served Sue Ambler, M.Ed well in her personal and professional life. Whether the mentor experience is face to face or Facebook Ms. Ambler attributes her success to those who have mentored her, those she has mentored and looks forward to the future mentors in her life.
About the Speaker
Sue Ambler, President and CEO of Workforce Development Council Snohomish County, has 30 years of experience in public and private industry throughout the greater Puget Sound area, including 16 years in public K-12 and higher education as both faculty and as a provider of career and academic advising, disability services coordination, outreach, recruitment, crisis management, youth advocacy, and curriculum and accreditation standards development.
Sue has been with the Workforce Development Council Snohomish County since 2003 and has served at the President and CEO from 2006 to present. In addition to Workforce Development duties she is active in the community and is committed to the economic vitality of the Puget Sound region. Some of her current activities include:
United Way Snohomish County Board and 2011 Community Campaign Chair Junior Achievement Snohomish County Board First Robotics Executive Advisory Board Washington Economic Development Council Snohomish County Board Economic Development District Board Puget Sound Regional Council South Snohomish County Chamber of Commerce Board Washington Workforce Association Board Investing in Families Steering Committee Snohomish County Blueprint 2015 Edmonds Community College Presidential Search Committee
Ms. Ambler has a Bachelor’s degree from Washington State University and Master’s degree in Education from University of Washington Bothell.
CSS Mock Technical Interview Workshop
Thursday, January 20, 2011
6:00 PM - 8:00 PM
UW1-310 (Windows Lab)
Software Development Engineer in Test
Microsoft Employee and CSS Alumni (’04), Denise Mosbrucker, will be presenting a CSS Mock Technical Interview Workshop. Learn secrets to Microsoft technical interviews, how to maximize your success in a technical interview, tips, etc.
Denise Mosbrucker has been working at Microsoft for the past two and half years. Currently she works as a Software Development Engineer in Test on Microsoft’s Surface team.
Wednesday, November 17, 2010
5:45pm - 6:45pm
WorldWide Windows Azure Technical Specialist
Just what on earth is Windows Azure? Why should you care about "the cloud"? How do you get started? What can you really do with Windows Azure? Please join us for this look at Microsoft's Windows Azure platform. Find out what it really is, how you can get started with the least amount of fuss and what you can really do with a few thousand compute nodes on demand.
About Mr. Aiken
David Aiken is currently a WorldWide Windows Azure Technical Specialist at Microsoft HQ. He has been working on Windows Azure since January 2008, well before there were bits - helped get earlier adopters onto the platform when deployment was still "send your code to thingy", and had a few customers featured at PDC 08 & 09. Now David spends his time showing customers how to build apps & services for the cloud. Before this, he worked on Visual Studio 2008 & .Net Framework 3.5 evangelism, producing demos such as DinnerNow and the Visual Studio training kit. Allegedly, he once claimed you could solve every problem with a single line of PowerShell script.