Home Page for Seth Rogers (Old Picture)
History
In the beginning, I was a lowly undergrad at Williams College. In 1991 I
became a graduate student at the
University of Michigan,
Department of Electrical Engineering and Computer Science, Intelligent Systems area. I
defended my thesis work on January 21, 1997.
After that, I worked at the the DaimlerChrysler Research
and Technology Center, Telematics
Research Group in sunny Palo Alto, until December 5, 2003.
Then, I worked at The Institute for the
Study of Learning and Expertise, affiliated with the Stanford Center for the Study of
Language and Information. My main project is a DARPA grant that
involves automatic learning of complex hierarchical tasks.
In March 2005 I moved to a startup called Circumnav Networks. There,
the plan was to connect the car to the Internet and provide some
useful services. We renamed ourselves to Dash Navigation and eventually
released a product called the Dash Express. The product was an
industry first in several ways, but sales weren't too good and we
eventually sold out to Research In Motion, where we were tasked with
replicating the best features of the Dash Express on the Blackberry
platform.
| Institution |
Degree |
Date |
Major |
| Williams College |
BA |
June 1991 |
Computer Science |
| University of Michigan |
PhD |
January 1997 |
Computer Science and Engineering |
Current Work
Crowd-sourced navigation and traffic solutions for the Blackberry smartphone.
Our current product is Blackberry Traffic, available on App World.
Former Work
Before Dash was acquired by Research In Motion, I built a traffic
procssing module to aggregate data from different sources into a single
model, offline processing tools to create historical model that is smooth
and complete with limited data, and a prediction library that estimates
travel times for arbitrary times and road segments.
At Stanford, I worked on a project to learn complex problem-solving
behavior such as game-playing, using the Icarus cognitive architecture.
At DC RTC, my research group is working on improving the driving
environment, with the ultimate goal of providing an option for a
future model of Mercedes cars. We are also working on automatic
digital map improvement using GPS data. See my papers in the
publications section.
I'm also doing contracting work for creating personalized software
interfaces to information or ecommerce systems. Contact me for
more information.
Thesis
For my thesis, I worked with John Laird and Paul
Nielsen in the Soar
group. My thesis research (abstract)
concerned symbolic function
approximation for controlling physical systems. The system uses the
Soar architecture for
general cognition.
I gave a talk on my research at
Stanford on March 12, 1997, at the Seminar on
Computational Learning and Adaptation. Here is the abstract.
Contacting Me
| Address |
| Dash Navigation, Inc. |
| 888 Ross Drive, Sunnyvale, CA 94089 |
|
| Phone |
408-543-2900 Ext. 1051 |
| Fax |
408-400-0939 (Mark my name) |
| Work Email |
srogers@dash.net |
| Private Email |
seth@sun-rogers.com |
Web Stuff
I worked on the following two web sites:
On-line publications (most recent first)
- Rogers, S. (2011). Traffic Quality in a Probe Vehicle Network.
Proceedings of the 90th Meeting of the Transportation Research Board,
Washington DC.
- Langley, P., Choi, D., & Rogers, S. (2009). Acquisition of
Hierarchical Reactive Skills in a Unified Cognitive Architecture. Cognitive
Systems Research, 10 316-332.
- Langley, P., Laird, J. E., & Rogers, S. (2009). Cognitive
Architectures: Research Issues and Challenges. Cognitive Science Research,
10, 141-160
- Langley, P., & Rogers, S. (2005). An Extended Theory of Human Problem
Solving. Proceedings of the 27th Annual Meeting of the Cognitive Science
Society. Stresa, Italy.
- Langley, P., & Rogers, S. (2004).
Cumulative learning of hierarchical skills (183K, 8 pages). Proceedings
of the Third International Conference on Development
and Learning. San Diego, CA.
- Stefan Schoredl, Kiri Wagstaff, Seth Rogers, Pat Langley,
Christopher Wilson, Mining GPS Traces for Map
Refinement (857K, 29 pages), Data Mining and Knowledge Discovery,
Volume 9, Issue 1, July 2004, pp. 59-87.
- Seth Rogers, Wenbing Zhang, Development
and Evaluation of a Curve Warning
System for Trucks (180K, 4 pages), IEEE IV2003 Intelligent Vehicles
Symposium Proceedings, IEEE Intelligent Transportation Systems
Council, June 2003, pp. 294-297.
- Kiri Wagstaff, Claire Cardie, Seth Rogers, and Stefan Schroedl, Constrained K-means Clustering with
Background Knowledge, (230K, 8 pages), Proceedings of the
Eighteenth International Conference on Machine Learning,
pp. 577-584, June 28-July 1, 2001.
- Seth Rogers, Claude-Nicolas Fiechter, Cynthia Thompson, Adaptive User Interfaces for Automotive
Environments (4408K, 6 pages), IEEE Intelligent Vehicles
Symposium, Dearborn, MI, October 2000, pp. 662-667. [PDF
(77K(!))]
- Seth Rogers, Creating and Evaluating
Highly Accurate Maps with Probe Vehicles (166K, 6 pages), IEEE Conference on Intelligent
Transportation Systems, Dearborn, MI, October 2000, pp. 125-130.
- Claude-Nicolas Fiechter and Seth Rogers, Learning Subjective Functions with Wide
Margins (4687K, 8 pages), Seventeenth International
Conference on Machine Learning, Stanford, CA, June 2000,
pp. 287-294.
- Claude-Nicolas Fiechter and Seth Rogers, Learning
Subjective Functions with Wide Margins
(4651K, 8 pages), AAAI Spring
Symposium on Adaptive User Interfaces, Stanford, CA, March 2000,
pp. 40-47.
- Christopher Pribe and Seth Rogers, Learning
to Associate Driver Behavior with Traffic Controls (43K, 14
pages), Transportation Research Record, No. 1679, November 1999,
pp. 95-100. [HTML]
- Seth Rogers, Pat Langley, Chris Wilson, Predicting
Lane Occupancy Using GPS and Digital
Maps (632K, 27 pages), submitted to the
Machine Learning special issue on Unsupervised Learning.
- Seth Rogers, Pat Langley, Chris Wilson, Mining
GPS Data to Augment Road Models (214K, 11
pages),
International Conference on Knowledge Discovery & Data Mining,
San Diego, August 1999, pp. 104-113. [HTML]
- Seth Rogers, Claude-Nicolas Fiechter, Pat Langley, An adaptive interactive agent for route advice
(6551K, 9 pages), Third
International Conference on Autonomous Agents, Seattle, WA, May
1999, pp. 198-205. [HTML]
- Seth Rogers, Claude-Nicolas Fiechter, A
Route
Advice Agent that Models Driver Preferences (4364K, 8 pages), AAAI Spring
Symosium on Agents with Adjustable Autonomy, Stanford, CA, March
1999, pp. 106-113. [HTML] [PDF]
- Christopher Pribe and Seth Rogers, Learning
to Associate Driver Behavior with Traffic Controls (43K, 14
pages), 78th Annual Meeting
of the Transportation Review Board, Washington, DC, January
1999. [HTML]
- Chris Wilson, Seth Rogers, Shawn Weisenburger, The Potential of Precision Maps in Intelligent
Vehicles (69K, 4 pages), 1998
IEEE
International Conference on Intelligent Vehicles, Stuttgart,
Germany, October 1998, pp. 419-422.
- Seth Rogers, Pat Langley, Personalized
Driving
Route Recommendations (258K, 5 pages), 1998 AAAI
Workshop on Recommender Systems, Madison, WI, July 1998,
pp. 96-100.
- Seth Rogers, Pat Langley, Interactive
Refinement of Route Preferences for Driving (273K, 5 pages), 1998
Stanford Spring Symposium Interactive and Mixed-Initiative
Decision-Theoretic Systems, Stanford, CA, March 1998, pp. 109-113.
- Seth Rogers, Simon Handley, Pat Langley, Modeling
Individual Driving Route Prefernces from
Relative Feedback (517K, 16 pages), submitted to Fifteenth International
Conference on
Machine Learning.
- Seth Rogers, Pat Langley, Brian Johnson and Annabel Liu, Personalization of the Automotive Information
Environment (233 KB, 6 pages), in R. Engels, B. Evans, J. Herrmann
and F. Verdenius (Eds.). Proceedings
of the workshop on Machine Learning in the real world; Methodological
Aspects and Implications. Nashville, TN, July 12th, 1997, pp.
27-33.
- Symbolic Performance & Learning in
Continuous-Valued Environments (3921KB, 176 pages), Dissertation.
Here is the abstract.
- Symbolic Performance & Learning in
Continuous-Valued Environments (624KB), submitted to the Fourteenth
International Conference on Machine Learning.
- Symbolic Performance & Learning in
Complex Environments (358KB), unpublished longer version of AAAI
abstract.
- Symbolic Performance & Learning in
Complex Environments (44KB), AAAI
1996 National Conference on Artificial Intelligence Student
Abstract and Poster Program, p. 1405. (new
version (161KB))
- New Results on Learning from Experience
in Continuous Domains (140KB), unpublished technical update.
- Increasing Learning Rate via
Active Goal Selection (86KB), in the 1995
AAAI
Symposium on Active Learning, pp. 95-98.
- Robert E. Wray III, Ronald Chong, Joseph Phillips, Seth Rogers,
William Walsh, John E. Laird: Organizing
Information in Mosaic: A Classroom Experiment (HTML). Computer
Networks and ISDN Systems 28(1&2): 167-178 (1995).
- Learning from Experience in
Continuous Domains (428KB), Thesis Proposal.
Other Seth Rogers on the web
It was a suprise to see how much of my life is documented on the
web, especially things I didn't know about. Apparently I am a:
- Hockey
player
- Rabbit
breeder
- Mentally ill
and
shot(!) (Strangely, this is a slightly different name)
- Flow
cytometrist (?)
- Kiosk enclosure
manufacturer
- And
more!
Go to ATL
Students Page
Go to ATL Home Page.
Send e-mail to
srogers@csli.stanford.edu.