I'm currently trying to find my way into the geometric description of Quantum Mechanics. I therefore started reading:

Geometry of state spaces. In: Entanglement and Decoherence (A. Buchleitner et al., eds.). Lecture Notes in Physics 768, Springer Verlag, Berlin, New York, 2009, 1-60.

A document that can also be found as a manuscript via: http://www.physik.uni-leipzig.de/~uhlmann/PDF/UC07.pdf

Even though I thought that I have a solid background in abstract algebra I somewhat got lost in Chapter 2 when he's trying to classify all the *-algebras that represent actual physical systems (starting at page 24 in the document).

Do you have some recommendations for texts that introduce the *-algebra language in Quantum Mechanics in a more 'detailed' way. Because I kind of have the feeling that at a certain point Uhlmann just keeps skipping steps and I also lack some of the physical intuition concerning partial traces, canonical traces, purification and all that. From time to time I'd also be happy to see a concrete example.


The original reference to Wedderburn's theorem is
J.H. Wedderburn, On hypercomplex numbers, Proceedings of the London Mathematical Society 2 (1908), 77-118.

For more on *-algebras, try Section 2.2 in Volume 3 and Section 1.4 of Volume 4 of Thirring, A course in mathematical physics.

Also of interest might be the following Wikipedia articles:
and more links in

  • $\begingroup$ Thirring's exposition looks nice and seems to be a nice start. But I think he's not talking about partial trace, purification, etc. $\endgroup$
    – MrLee
    Oct 1 '12 at 22:11
  • $\begingroup$ @MrLee: A partial trace is simply a trace over some factors of a tensor product Hilbert space. I don't know what purification is; it doesn't seem to be a standard concept. $\endgroup$ Oct 2 '12 at 8:09
  • $\begingroup$ @ArnoldNeumaier purification is that every mixed state can be reprensented as the partial trace of a pure state living in a higher dinensional space. (Not sure if this works in thr c* algebra) $\endgroup$
    – lalala
    Apr 1 '18 at 11:39
  • $\begingroup$ @lalala: Thanks for the explanation. The C^* analogue is the GNS construction, which represents each state as a pure state in some Hilbert space. $\endgroup$ Apr 1 '18 at 14:36

OP wrote(v2):

I also lack some of the physical intuition concerning partial traces, canonical traces, purification and all that.

Apart from the mathematical subject of *-algebras, I get the impression that OP really wants to study quantum information rather than quantum mechanics. If this hunch is correct, then I can recommend for starters from the physics side, the textbook

M. Nielsen and I.L Chuang, Quantum Computation and Quantum Information, Cambridge University Pres (2000).


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.