<Chapter>
<Heading>Projective resolutions and the bounded derived category</Heading>
What is implemented so far for working with the bounded derived
category <Math>\mathcal{D}^{b}( \modc A )</Math>. We use the
isomorphism <Math>\mathcal{D}^{b}( \modc A ) \cong
\mathcal{K}^{-,b}(\proj A)</Math>, and will hence need a way to
describe complexes where all objectives are projective (or, dually,
injective).
<Section>
<Heading>Projective and injective complexes</Heading>
<ManSection>
<Oper Name="InjectiveResolution" Arg="M" />
<Description>
Arguments: <Arg>M</Arg> -- a module.<Br />
</Description>
<Returns>The injective resolution of <Arg>M</Arg> with <Arg>M</Arg>
in degree <Math>-1</Math>.
</Returns>
<Description>
</Description>
</ManSection>
<ManSection>
<Prop Name="IsProjectiveComplex" Arg="C" />
<Description>
Arguments: <Arg>C</Arg> -- a complex.<Br />
</Description>
<Returns>true if <Arg>C</Arg> is either a finite complex of
projectives or an infinite complex of projectives constructed as a
projective resolution (<Ref Func="ProjectiveResolutionOfComplex"/>),
false otherwise.
</Returns>
<Description>A complex for which this property is true, will be
printed in a different manner than ordinary complexes. Instead of
writing the dimension vector of the objects in each degree, the
indecomposable direct summands are listed (for
instance <C>P1</C>, <C>P2</C> … , where <Math>P_i</Math> is the
indecomposable projective module corresponding to
vertex <Math>i</Math> of the quiver). Note that if a complex is both
projective and injective, it is printed as a projective complex.
</Description>
</ManSection>
<ManSection>
<Prop Name="IsInjectiveComplex" Arg="C" />
<Description>
Arguments: <Arg>C</Arg> -- a complex.<Br />
</Description>
<Returns>true if <Arg>C</Arg> is either a finite complex of
injectives or an infinite complex of injectives constructed
as <Math>D\mathrm{Hom}_{A}(-,A)</Math> of a projective complex
(<Ref Func="ProjectiveToInjectiveComplex"/>), false otherwise.
</Returns>
<Description>A complex for which this property is true, will be
printed in a different manner than ordinary complexes. Instead of
writing the dimension vector of the objects in each degree, the
indecomposable direct summands are listed (for
instance <C>I1</C>, <C>I2</C> … , where <Math>I_i</Math> is the
indecomposable injective module corresponding to
vertex <Math>i</Math> of the quiver). Note that if a complex is both
projective and injective, it is printed as a projective complex.
</Description>
</ManSection>
<ManSection>
<Oper Name="ProjectiveResolution" Arg="M" />
<Description>
Arguments: <Arg>M</Arg> -- a module.<Br />
</Description>
<Returns>The projective resolution of <Arg>M</Arg> with <Arg>M</Arg>
in degree <Math>-1</Math>.
</Returns>
<Description>
</Description>
</ManSection>
</Section>
<Section>
<Heading>The bounded derived category</Heading>
Let <Math>\mathcal{D}^{b}( \modc A )</Math> denote the bounded
derived category. If <Math>C</Math> is an element
of <Math>\mathcal{D}^{b}( \modc A )</Math>, that is, a
bounded complex of <Math>A</Math>-modules, there exists a projective
resolution <Math>P</Math> of <Math>C</Math> which is a complex of
projective <Math>A</Math>-modules quasi-isomorphic
to <Math>C</Math>. Moreover, there exists such a <Math>P</Math> with
the following properties:
<List>
<Item><Math>P</Math> is minimal (in the homotopy category).</Item>
<Item><Math>C</Math> is bounded, so <Math>C_i = 0</Math> for <Math>i
< k</Math> for a lower bound <Math>k</Math> and <Math>C_i =
0</Math> for <Math>i > j</Math> for an upper
bound <Math>j</Math>. Then <Math>P_i = 0</Math> for <Math>i <
k</Math>, and <Math>P</Math> is exact in degree <Math>i</Math>
for <Math>i > j</Math>.</Item>
</List>
The function <C>ProjectiveResolutionOfComplex</C> computes such a
projective resolution of any bounded complex. If <Math>A</Math> has
finite global dimension, then <Math>\mathcal{D}^{b}( \modc A )</Math>
has AR-triangles, and there exists an algorithm for computing the
AR-translation of a complex <Math>C \in \mathcal{D}^{b}( \modc A
)</Math>:
<List>
<Item>Compute a projective resolution <Math>P'
of <Math>C</Math>.</Item>
<Item>Shift <Math>P' one degree to the right.
<Item>Compute <Math>I = D\mathrm{Hom}_{A}(P',A) to get a
complex of injectives.</Item>
<Item>Compute a projective resolution <Math>P</Math> of <Math>I</Math>.</Item>
</List>
Then <Math>P</Math> is the AR-translation of <Math>C</Math>, sometimes
written <Math>\tau(C)</Math>.
The following documents the <Package>QPA</Package> functions for
working with complexes in the derived category.
<ManSection>
<Oper Name="ProjectiveResolutionOfComplex" Arg="C" />
<Description>
Arguments: <Arg>C</Arg> -- a finite complex.<Br />
</Description>
<Returns>A projective complex <Math>P</Math> which is the projective
resolution of <Math>C</Math>, as described in the introduction to
this section.
</Returns>
<Description>If the algebra has infinite global dimension, the
projective resolution of <Math>C</Math> could possibly be infinite.
</Description>
</ManSection>
<ManSection>
<Oper Name="ProjectiveToInjectiveComplex" Arg="P" />
<Oper Name="ProjectiveToInjectiveFiniteComplex" Arg="P" />
<Description>
Arguments: <Arg>P</Arg> -- a bounded below projective complex.<Br />
</Description>
<Returns>An injective complex <Math>I = D\mathrm{Hom}_{A}(P,A)</Math>.
</Returns>
<Description><Math>P</Math> and <Math>I</Math> will always have the
same length. Especially, if <Math>P</Math> is unbounded above, then
so is <Math>I</Math>. If <Math>P</Math> is a finite complex (that
is; <C>LengthOfComplex(P)</C> is an integer) then the simpler method
<C>ProjectiveToInjectiveFiniteComplex</C> is used.
</Description>
</ManSection>
<ManSection>
<Oper Name="TauOfComplex" Arg="C" />
<Description>
Arguments: <Arg>C</Arg> -- a finite complex over an algebra of
finite global dimension.<Br />
</Description>
<Returns>A projective complex <Math>P</Math> which is the
AR-translation of <Arg>C</Arg>.
</Returns>
<Description>This function only works when the algebra has finite
global dimension. It will always assume that both the projective
resolutions computed are finite.
</Description>
</ManSection>
<Subsection>
<Heading>Example</Heading>
The following example illustrates the above mentioned functions and
properties. Note that both <C>ProjectiveResolutionOfComplex</C>
and <C>ProjectiveToInjectiveComplex</C> return complexes with a
nonzero <E>positive</E> part, whereas <C>TauOfComplex</C> always
returns a complex for which <C>IsFiniteComplex</C> returns true. Also
note that after the complex <C>C</C> in the example is found to have
the <C>IsInjectiveComplex</C> property, the printing of the complex
changes. <P />
The algebra in the example is <Math>kQ/I</Math>, where <Math>Q</Math>
is the quiver <Math>1 \longrightarrow 2 \longrightarrow 3</Math>
and <Math>I</Math> is generated by the composition of the arrows. We
construct <Math>C</Math> as the stalk complex with the
injective <Math>I_1</Math> in degree 0.
<ManSection>
<Oper Name="StarOfMapBetweenProjectives" Arg="f, list_i, list_j" />
<Oper Name="StarOfMapBetweenIndecProjectives" Arg="f, i, list_j" />
<Oper Name="StarOfMapBetweenDecompProjectives" Arg="f, list_i, list_j" />
<Description>
Arguments: <Arg>f</Arg> -- a map between to projective modules
<Math>P = \bigoplus P_i</Math> and <Math>Q = \bigoplus Q_j</Math>,
each of which were constructed as direct sums of indecomposable
projective modules; <Arg>list_i</Arg> -- describes the summands of
<Math>P</Math>; <Arg>list_j</Arg> -- describes the summands of
<Math>Q</Math>. If <Math>P = P_1 \oplus P_3 \oplus P_3</Math>
(where <Math>P_i</Math> is the indecomposable projective
representation in vertex <Math>i</Math>), then <Arg>list_i</Arg>
is [1,3,3].<Br />
</Description>
<Returns>The map <Math>f^* = \Hom_A(f,A): \Hom_A(Q,A) \rightarrow \Hom_A(P,A)</Math>
in <Math>A^{\mathrm{op}}</Math> (where <Math>A</Math> is the
original algebra).
</Returns>
<Description>The function <C>StarOfMapBetweenProjectives</C> is supposed to
be called from within the <C>ProjectiveToInjectiveComplex</C> method, and
might not do as expected when called from somewhere else. <P />
The other similarly named functions are called from within the first.
</Description>
</ManSection>
</Section>
</Chapter>
¤ Dauer der Verarbeitung: 0.15 Sekunden
(vorverarbeitet)
¤
Die Informationen auf dieser Webseite wurden
nach bestem Wissen sorgfältig zusammengestellt. Es wird jedoch weder Vollständigkeit, noch Richtigkeit,
noch Qualität der bereit gestellten Informationen zugesichert.
Bemerkung:
Die farbliche Syntaxdarstellung ist noch experimentell.
