Thus spake Rich Murray circa 10-02-26 09:18 AM: > http://www.wired.com/magazine/2010/02/ff_algorithm/all/1
Thanks for forwarding this. On the surface, it seems like it's a particular type of genetic algorithm. The sparsity function is the objective, the low-res measurements are a constraint, the shapes that are plugged into the holes are the terminals, and the methods for choosing when to place smaller shapes are the operators. The population consists of individual, filled-in images. Given that, I wonder if CS has the same persnickety implementation issues that GAs have? TANSTAAFL, I mean. I.e. the magic lies in getting the objective function, terminals, and operators just right to make any progress. And behind every "miraculous" success lies a long evolution of tweak a little, here, tweak a little there, etc. Anyone know more about this and can shatter my analogy before I attempt to wade through a bunch of CS/l_1 papers? -- glen e. p. ropella, 971-222-9095, http://agent-based-modeling.com ============================================================ FRIAM Applied Complexity Group listserv Meets Fridays 9a-11:30 at cafe at St. John's College lectures, archives, unsubscribe, maps at http://www.friam.org
