Nonlinear System Identification: From Classical Approaches to Neural Networks and Fuzzy Models Book

Extremely good book, especially due to many explicit computations. Some notations are a little awkward, e.g. calling a one-form a vector (although confusion can be remedied since we have the metric). The exposition on chapter 2 (congruences), chapter 3 (hypersurfaces) chapter 4 (Hamiltonian formalism) and chapter 5 (black holes) are extremely illuminating due to explicit computations presented (which is often missing from Wald). Black hole thermodynamics are covered pretty well. Thus this is a very good text to complement Wald.

To reap most benefits, one should be acquainted with basics of general relativity (at the level of Schutz's introductory text) and perhaps some skimming through of more advanced text like Wald or Carroll's GR texts (because some references to the more advanced ideas are made, e.g. congruences). And one should be acquainted with some basic notions of differential geometry as well, because without it even this text may appear harder than it is supposed to be. Indeed more elegance can be found in Wald or Hawking or other texts, however this bridges the gap in technicality very well.

One should really not treat this text as a source to learn GR. One should use this to complement other texts.