|
|
|
|
|
|
Help | Seminars List | Add Seminar | Edit Seminars | Tips for organisers | RSS | ics Calendar | Search | Send comments about this website to seminar-master@cecs.anu.edu.au
Contact: Michelle.Moravec@anu.edu.au CS PHD MONITORING
What we know about what we don't know -- Can we marry model assessment with universal induction?Mr Ian Wood (School of Computer Science, CECS)DATE: 2009-09-25 TIME: 14:00:00 - 14:30:00 LOCATION: RSISE Seminar Room, ground floor, building 115, cnr. North and Daley Roads, ANU ABSTRACT: Inductive reasoning is the fundamental technique we use to make sense of the world. In science and machine learning, we create models and verify them against real world data. Generally, we first attempt to find a 'best fit' from among our models, then assess the ability of our selected model to predict future data. In machine learning literature, this second step is called "model assessment". Universal or Solomonoff induction is a theoretical (though incalculable) ideal inductive procedure. I attempt to look into model assessment from a universal perspective (in the Solomonoff sense). This leads me to a novel perspective on characterisation of universal Turing machines.
|