[Nrg-l] FW: Statistics and Networks Course

Mark Crovella crovella at cs.bu.edu
Fri Aug 26 12:01:41 EDT 2005

This will be a good course!


-----Original Message-----
From: Eric Kolaczyk [mailto:kolaczyk at math.bu.edu] 
Sent: Friday, August 26, 2005 12:00 PM
To: lgrosser at bu.edu; wckarl at bu.edu; Mark Crovella; byers at cs.bu.edu;
robins at hsph.harvard.edu; best at bu.edu; matta at cs.bu.edu; delisi at bu.edu;
Timothy Gardner; kasif at bu.edu; Sucharita Gopal; redner at bu.edu; hes at bu.edu;
brown at neurostat.mgh.harvard.edu
Subject: Statistics and Networks Course


Apologies for the mass-sending on this email.  This is just a (one-time!)
notice regarding a course I'll be teaching this fall, on statistics and
networks.  Course is aimed at graduate students in quantitative areas
involving modeling and analysis of networks and network data, including
bioinformatics, computer science, engineering, math/stat, physics, etc.

A summary of relevant info is below.  More can be found in the syllabus at
http://math.bu.edu/people/kolaczyk/ma881.html .

Please pass on to interested students and colleagues.  Thanks!



COURSE: CAS MA 881 -- Statistics for the Network Sciences
TIME: Mondays, 3-6pm
LOCATION: GCB 208 (above the Guitar Center, at 750 Comm Ave)


1. A First Look at Network Graphs
    -- Background on graphs; examples of comm-, bio-, and social-nets.
2. Descriptive Analysis of Network Graphs
    -- Visualization; quantitative summaries (e.g., size, degree,
      assortativity, centrality, mutuality, others) 3. Probabilistic Models
for Network Graphs
    -- Graph ensembles; Erdos-Renyi; Small-worlds; Heavy-Tails; copying;
4. Implications of Network Structure
    -- Hubs and authorities; vulnerability; communities; epidemics.
5. Inference of Network Graph Structure
    -- Exponential models (classical, max-ent); Gibbs models; graphical
6. Testing for Topological Motifs
    -- Parametric tests; enumerative and sampling-based methods.
7. Impact of Sampling on Network Graph Inference
    -- Error & bias; heavy-tailed or not; species problems.
8. Analysis of Network Indexed Data
    -- Descriptive analysis; dimensionality reduction; regression
       and classification (e.g., kernel methods, MRFs, graphical
9. Network Tomography and Related Topics
    -- Traffic matrix estimation; network kriging; etc.
10. Analysis of dynamic networks.

More information about the Nrg-l mailing list