OREGON STATE UNIVERSITY
Probabilistic and decision-theoretic planning, which operates under conditions of uncertainty, has important applications in science and engineering, but such stochastic methods have been under-utilized because these planners do not scale to large, complicated problems.
In contrast, the past decade has seen tremendous scale-up in deterministic planning techniques, which do not directly address the same challenges of uncertainty head on. This project provides a systematic framework for exploiting this progress in deterministic planning technology - be it classical, temporal or partial satisfaction planning problems - in stochastic planning as well, by novel adaptation of theory and/or methods of determinization, hindsight optimization, and machine learning.
This project is helping to bridge the traditional divide between deterministic and stochastic planning communities, and it is bringing stochastic planning technology to real-world applications.
Choose a quarter and click "Go."
| AWARD OVERVIEW |
| Award Number |
0905678 |
Funding Agency |
National Science Foundation |
| Total Award Amount |
$278,793 |
Project Location - City |
Corvallis |
| Award Date |
06/22/2009 |
Project Location - State |
OR |
| Project Status |
More than 50% Completed |
Project Location - Zip |
97331-8507
|
| Jobs Reported |
1.07 |
Congressional District |
04 |
| Project Location - Country |
US |
|
|
Recipient Information
(Grants)
| Recipient Information (Grants) |
|
Recipient Name
|
OREGON STATE UNIVERSITY |
| Recipient DUNS Number |
053599908
|
| Recipient Address |
KERR ADMINISTRATION B306 |
| Recipient City |
CORVALLIS |
| Recipient State |
Oregon |
| Recipient Zip |
97331 |
| Recipient Congressional District |
04 |
| Recipient Country |
USA |
Required to Report Top 5
Highly Compensated Officials |
No |
Projects and Jobs Information
| Projects and Jobs Information |
| Project Title |
RI: Medium: Collaborative Research: Solving Stochastic Planning Problems Through Principled Determinization |
| Project Status |
More than 50% Completed |
| Final Project Report Submitted |
No |
| Project Activities Description |
Mathematics |
| Quarterly Activities/Project Description |
The grant supported the research activities of two Graduate Research Assistants (GRAs).
One GRA continued worked on formulating speedup learning algorithms for stochastic planning. He completed his work on the basic infrastructure and ran preliminary experiments. The results show that the speedup learning algorithm is able to significantly speedup the planning time of the UCT planner in two test domains. Further work will continue to refine and test the algorithm as well as analyze it at a theoretical level.
A second GRA worked on Monte-Carlo planning for stochastic games in a policy switching framework. He completed the camera ready version of our policy switching paper, which will be appearing at the 2013 International Conference on Automated Planning and Scheduling. He has also completed the implementation and testing of a new policy switching algorithm in the domain of Galcon Fusion, an abstract real-time strategy game. The initial tests show that the basic version of Galcon is too simple to allow for the potential advantages of switching to be shown. He has since designed and implemented a more complex version of Galcon and will be testing the algorithm in the coming weeks. He also began working on his Doctor of Philosophy (PhD) qualifier exam this quarter and will complete that in early April. |
| Jobs Created |
1.07 |
| Description of Jobs Created |
The funds for the project have created a new graduate research assistant (GRA) position(s) in my group. The funds provide for the salary, benefits, and tuition. The GRA has now graduated and the funds have since been used to support the work of a number of Doctor of Philosophy (Ph.D.) students. |
Purchaser Information
(Grants)
| Purchaser Information |
| Contracting Office ID |
Not Reported |
| Contracting Office Name |
Not Available |
| Contracting Office Region |
Not Available |
| TAS Major Program |
49-0101 |
| Award Information |
| Award Date |
06/22/2009 |
| Award Number |
0905678 |
| Order Number |
|
| Award Type |
Grants |
| Funding Agency ID |
49 |
| Funding Agency Name |
National Science Foundation |
| Funding Office Name |
Not Available |
| Awarding Agency ID |
49 |
| Awarding Agency Name |
National Science Foundation |
| Amount of Award |
$278,793 |
| Funds Invoiced/Received |
$276,231 |
| Expenditure Amount |
$278,793 |
| Infrastructure Expenditure Amount |
$0 |
| Infrastructure Purpose and Rationale |
Not Reported |
| Infrastructure Point of Contact Name |
Not Reported |
| Infrastructure Point of Contact Email |
Not Reported |
| Infrastructure Point of Contact Phone |
Not Reported |
| Infrastructure Point of Contact Address |
Not Reported |
| Infrastructure Point of Contact City |
Not Reported |
| Infrastructure Point of Contact State |
Not Reported |
| Infrastructure Point of Contact Zip |
Not Reported |
Product or Service Information
(Grants)
| Product or Service Information |
| Primary Activity Code |
U03.05 |
| Activity Description |
Mathematics |
| Sub-Awards Information |
| Sub-awards to Organizations |
0 |
| Sub-award Amounts to Organizations |
$0 |
| Sub-Awards to Individuals |
0 |
| Sub-Award Amounts to Individuals |
$0 |
| Number of Sub-awards less than $25,000/award |
0 |
| Amount of Sub-awards less than $25,000/award |
$0 |
| Number of payments to vendors greater than $25,000 |
0 |
| Total Amount of payments to vendors greater than $25,000/award |
$0 |
| Number of payments to vendors less than $25,000/award |
18 |
| Total Amount of payments to vendors less than $25,000/award |
$22,396 |
| Location Information |
| Latitude, Longitude |
44º 33' 53",
-123º 16' 44" |
| Congressional District |
04 |
| Address 1 |
Oregon State University |
| Address 2 |
|
| City |
Corvallis |
| County |
Benton |
| State |
OR |
| Zip |
97331-8507 |
|
 |