Minutes for UA MOPS meeting, March 9, 2006
In attendance:
Kobus Barnard, Alon Efrat, Bongki Moon, Scott Morris, Praveen Rao
Main points:
1. Minutes will continue to be done by Kobus for a while. Someone else will
pick it up on a "rotating" basis.
2. There will be no meeting next week (UA spring break).
3. Scott reported some success wrapping OrbFit differential calculations.
Failures still occur that need to be examined as too whether they are due
to the underlying code, or the problems wrapping this piece of software.
4. Technical discussion focused on robustness testing of the matching
process. Matching gives us maximum likelihood under some assumptions, but
can contain many incorrect matches. We discussed heuristic proposed by
Alon to evaluate robustness of matches. Heuristics are needed because we
are evaluating probabilities subject to 1-1 matching.
We also discussed (but did not make a lot of progress on) the extent to
which plausible but incorrect matches will cause problems. Our intuition
is that a correct orbit can collect a number of incorrect detections that
are consistent with the orbit without incurring a lot of error. However, a
more precise formulation should be sought.
--
Kobus Barnard
[EMAIL PROTECTED]
http://kobus.ca
730 Gould Simpson, U. Arizona (520-621-4237)
_______________________________________________
LSST-data mailing list
[email protected]
http://www.lsstmail.org/mailman/listinfo/lsst-data
