Skip to main content
16 events
when toggle format what by license comment
Jul 24, 2019 at 8:50 vote accept Praphulla Koushik
Jul 24, 2019 at 6:32 history edited Praphulla Koushik CC BY-SA 4.0
added 193 characters in body
Jul 24, 2019 at 1:28 comment added Praphulla Koushik @FernandoMuro thank you. I will see that...
Jul 23, 2019 at 18:26 comment added Fernando Muro See the motivation section in en.m.wikipedia.org/wiki/Lie_algebra_cohomology
Jul 23, 2019 at 18:05 comment added Praphulla Koushik @FernandoMuro I mean manifold cohomology only... I do not know why it seems like this is not what I want.. I am looking for some easy way to compute using the Lie group structure.. I was thinking if I can take some subcomplex and still get cohomology of manifold... I will read about Chevalley-Eilenberg complex. Thanks,. It looks like you see something is not ok in my question.. Please let me know what it is.. I can explain or correct myself..
Jul 23, 2019 at 18:00 comment added Praphulla Koushik I see one downvote.. if there is anything I can improve in the question, do let me know.. I may be wrong..
Jul 23, 2019 at 17:59 comment added Fernando Muro @praphullakoushik if you mean the manifold cohomology you have the trivial answer: take the whole complex. I guess this is not what you want. As pointed above, under some hypotheses (simply connected, etc.) the Chevalley-Eilenberg complex of its Lie algebra also computes the same cohomology and it has finite dimension.
Jul 23, 2019 at 17:28 comment added Praphulla Koushik @MikeMiller I did not think of the possible confusion... If I say cohomology of the Lie group $G$ it is not clear if I am saying about cohomology of the underlying manifold or the cohomology of topological space $BG$.. Here I mean not $BG$. Are there other possible interpretation of “cohomology of Lie group G”
Jul 23, 2019 at 17:13 comment added Praphulla Koushik @FernandoMuro I mean the cohomology of underlying manifold.. I am looking for some easy way to compute the cohomology.. sorry for the confusion... is it ok now?
Jul 23, 2019 at 17:09 comment added Fernando Muro Could you clarify what notion of Lie group cohomology you're talking about?
Jul 23, 2019 at 15:25 comment added mme Wait, I am confused. When you say "cohomology of this Lie group", do you intend to mean the cohomology of the classifying space BG? You can't possibly extract this from a subcomplex of the de Rham complex because $H^*(BG)$ is nonzero in arbitrarily large degrees (and in particular past $\dim G$). If what you mean is that you want a nice complex computing cohomology of the underlying manifold then Gael's complex does what you want.
Jul 23, 2019 at 14:46 answer added Gael Meigniez timeline score: 7
Jul 23, 2019 at 14:13 comment added Praphulla Koushik @YCor if it is already written somewhere, can you please suggest reference.. I will read :)
Jul 23, 2019 at 14:02 comment added YCor When $G$ is compact a suitable subcomplex consists of those $G$-invariant forms. When $G$ is arbitrary (connected), maybe the subcomplex of $K$-invariant forms, for $K$ maximal compact subgroup of $G$, works, but I haven't checked.
Jul 23, 2019 at 14:00 history edited YCor CC BY-SA 4.0
added 2 characters in body
Jul 23, 2019 at 13:58 history asked Praphulla Koushik CC BY-SA 4.0