Definition 1
A trialgebra
with
and
associative
products on a vector space
(where
may be partially defined, only)
and
a coassociative coproduct on
is given if both
and
are bialgebras and the following compatibility
condition between the products is satisfied for arbitrary elements
:
|
|
|
whenever both sides are defined.
Definition 3.3.1
.
Left unit operation
: Compute the value
, bind
to the result and compute
; the result is the same as
with value
substituted for variable
.
|
|
|
Right unit operation
: Compute
, bind the result to
and returns
; the result is the same as
.
|
|
|
Associativity
: Compute
, bind the result to
, compute
, bind the result to
and compute
; the order of parentheses is irrelevant.
|
|
|
Definition 1.1
.
Let
be a commutative ring. A
-module
equipped with two associative unital
-algebra structures
and
is called a double algebra over
if the following properties hold:
-
A1.
-
A2.
-
A3.
-
A4.
-
A5.
-
A6.
-
A7.
-
A8.
for all
.
Definition 19
.
A degenerate elliptic bigrid of type
is a bigrid given by the formula
|
|
|
(7.17)
|
where
,
,
,
, and
is a degree two function on
of the form
|
|
|
(7.18)
|
with
,
,
,
,
.
Definition 1.5
.
[
M
, Definition 2.14]
.
A
matched pair of groupoids
is a pair of groupoids
over
with
denoted vertically and
horizontally, endowed with a left action
of
on
, and a right action
of
on
, satisfying
(1.7)
|
|
|
|
(1.8)
|
|
|
|
(1.9)
|
|
|
|
for all
,
such that the compositions
are possible.
Definition 4.6
.
Let us first say that a
quiver
in a category
is a pair of arrows
in
.
A
double quiver
is a
quiver in the category
of all quivers. That is, in
the “vertical and horizontal” notation, a double quiver is a pair
of morphisms of quivers
, where
and
are quivers in the usual sense:
,
, and
should preserve
:
(4.5)
|
|
|
|
In short, a double quiver is a collection of sets and maps
|
|
|
satisfying (
4.5
). By abuse of notation we shall say
that
is a double quiver over
; or
alternatively that
is a double quiver with sides in
and
; or that
is a double quiver with sides in
in
case
. An element
of
is called an
oriented box
and depicted as a box
|
|
|
where
,
,
,
,
and the four corners are
,
,
,
. In
this picture, we keep in mind the orientations top-to-bottom and
left-to-right. Morphisms of double quivers, or of double quivers
over
, are defined in the standard way.
Let
be a set and
,
be quivers over
denoted vertically and horizontally, respectively. The
coarse double quiver
with sides in
and
is the
collection
where
is the set of all quadruples
with
,
such that
(4.6)
|
|
|
|
Such a quadruple is called a
face
. We omit the obvious
description of the arrows.
If
is a double quiver over
then
there are maps
,
given by
|
|
|
Clearly
is a morphism of double quivers.
We shall say that
is
thin
if
is injective (any box is determined by its sides) and
is surjective.
We shall say that
is
vacant
if
is bijective.
Definition 5.4
.
The
extended Mukai lattice
is defined to be
with the pairing
|
|
|
where
stands for the cup product. The extended Mukai lattice
has a natural weight 2 Hodge structure
|
|
|
Definition 1 1
Let
,
.
We note
|
|
|
to say that we pass from
to
by eliminating one generator.This last is said to be a 1-sequence.
So for any
in
,the k-sequence
|
|
|
means that we pass from
to
by eliminating
generators,i.e there
exists
positive words
with
such as
|
|
|
|
|
|
|
|
|
Definition 2.1
.
Let
be homomorphisms of commutative
-algebras.
We say that
and
are
homotopic
(
)
if there exists a
-algebra homomorphism
such that
|
|
|
The map
is called a
homotopy
between
and
.
Let
be the equivalence relation generated by
.
The homomorphism
is a
homotopy equivalence
if
there exists a ring homomorphism
with
and
.