![[Back]](/images/prevpage.gif)
![[Index]](/images/index.gif)
![[Help]](/images/help.gif)
![[MSI]](/images/msi.gif)
![[ANU Online]](/images/online.gif)
Research Report MRR00-001
Virtual triangles
Avishay Vaknin
Abstract:
Motivated by Neeman's K-theory for triangulated
categories, we are studying virtual triangles in a triangulated
category. A virtual triangle is a triangle, which is a direct
summand of an exact triangle, where the other summand is a direct
sum of trivial triangles. We define exact triangle to be a
triangle, which any two of its maps belong to a distinguished
triangle, and a distinguished triangle on an identity map to be a
trivial triangle. A direct sum of trivial triangles is called a
splitting triangle, and a direct summand, where the other summand
splits, is called an easy direct summand. So, by definition,
virtual triangles are closed under easy direct summands. It turns
out that they are also closed under direct sums, mapping cones,
and the t-structure truncation operation. This leads, using
Neeman's methods, to the virtual K-theory for triangulated
categories. We have in this theory, a version of the theorem of
the heart, and it contains Quillen's K-theory for the heart as a
retract. Unlike Neeman's theory, we don't know if this two
K-theories coincide. The benefit of this theory is, that it is
functorial, and does not make any use of models, in the sense of
Waldhausen. We explain when pseudo triangles are all virtual
triangles, where a pseudo triangle is a triangle, that goes to a
long exact sequence by any homology or cohomology functor. At the
end, we introduce the notion of fuzzy triangles, in order to
construct a virtual triangle, which is not exact. This shows, that
exact triangles are not closed under easy direct summands.
This service is maintained by the
Mathematical Sciences Institute (MSI)
Comments to
webmaster@maths.anu.edu.au
URL: http://wwwmaths.anu.edu.au/