I think I should attend this talk (see below) by the renowned Vladimir Vapnik, but once again the language of computer science leaves me baffled:

EMPIRICAL INFERENCE SCIENCE

1. With appearance of computers the concept of natural science, its methodology, and its philosophy started a painful process of a paradigm change : From the concepts, methodology, and philosophy of a Simple World it moves to very different concepts, philosophy and methodology of a Complex World.
2. In such changes an important role belongs to the mathematical facts that were discovered by analysing the “Drosophila fly” of cognitive science the “Pattern recognition problem” and attempts to obtain their philosophical interpretation.
3. The results of these analyses lead to methods that go beyond classical concept of science: creating generative models of events explain-ability of rules principles of refutation.
4. Instead the new paradigm introduces direct search for solution (transductive inference, instead of inductive), meditative principle of decision making, unity of two languages for pattern description: technical (rational) and holistic (irrational). This lead to convergence of the exact science with humanities.
5. The goal of this talk is to demonstrate that attempts to obtain a better generalization using limited number of observations lead to non-classical paradigm of inference.
6. New paradigm of inference based on philosophy appropriate for the Complex World which requires reconsideration of:goals of science, methodology of science, and concepts of justification of science
7. In this talk I will consider Pattern recognition problem for which in many cases one can obtain irrefragable answers. However the philosophy of generalization developed for this problem is true for more general situations.

The talk is this Friday 3pm in Mudd 633.