SOM Assisted PDF Fitting

«   »
Su Mo Tu We Th Fr Sa
28 29 30 1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31 1
Date/Time:Thursday, 30 Apr 2009 from 4:10 pm to 5:10 pm
Location:A410 Zaffarano Hall
Phone:515-294-6952
Channel:Nuclear Physics
Actions:Download iCal/vCal | Email Reminder
Heli Honkanen (Iowa State University)

In this talk we discuss an exploratory Parton Distribution Function (PDF) fitting technique based on Self-Organizing Maps (SOMs). SOMs are a class of clustering algorithms based on competitive learning
among spatially-ordered neurons.

Heli Honkanen (Iowa State University)

In this talk we discuss an exploratory Parton Distribution Function (PDF) fitting technique based on Self-Organizing Maps (SOMs).SOMs are a class of clustering algorithms based on competitive learning
among spatially-ordered neurons. Our SOMs are trained on selections of stochastically generated PDF samples. The selection criterion for every optimization iteration is based on the features of the clustered PDFs. The idea of our method is to create means for introducing ``Researcher Insight'' instead of ``Theoretical bias''. In other words, we want to give up fully automated fitting procedure and eventually develop an interactive fitting program which would allow us to ``take the best of both worlds'', to combine the best features of both the standard functional form approach and the neural network approach.