text/deligne.tex
changeset 449 ae5fd0a7a8a3
parent 437 93ce0ba3d2d7
child 450 56a31852242e
equal deleted inserted replaced
448:c3c8fb292934 449:ae5fd0a7a8a3
     1 %!TEX root = ../blob1.tex
     1 %!TEX root = ../blob1.tex
     2 
     2 
     3 \section{Higher-dimensional Deligne conjecture}
     3 \section{Higher-dimensional Deligne conjecture}
     4 \label{sec:deligne}
     4 \label{sec:deligne}
     5 In this section we 
     5 In this section we prove a higher dimensional version of the Deligne conjecture
     6 sketch
     6 about the action of the little disks operad on Hochschild cochains.
     7 \nn{revisit ``sketch" after proof is done} 
       
     8 the proof of a higher dimensional version of the Deligne conjecture
       
     9 about the action of the little disks operad on Hochschild cohomology.
       
    10 The first several paragraphs lead up to a precise statement of the result
     7 The first several paragraphs lead up to a precise statement of the result
    11 (Theorem \ref{thm:deligne} below).
     8 (Theorem \ref{thm:deligne} below).
    12 Then we sketch the proof.
     9 Then we give the proof.
    13 
    10 
    14 \nn{Does this generalization encompass Kontsevich's proposed generalization from \cite[\S2.5]{MR1718044}, 
    11 \nn{Does this generalization encompass Kontsevich's proposed generalization from \cite[\S2.5]{MR1718044}, 
    15 that (I think...) the Hochschild homology of an $E_n$ algebra is an $E_{n+1}$ algebra? -S}
    12 that (I think...) the Hochschild homology of an $E_n$ algebra is an $E_{n+1}$ algebra? -S}
    16 
    13 
    17 %from http://www.ams.org/mathscinet-getitem?mr=1805894
    14 %from http://www.ams.org/mathscinet-getitem?mr=1805894
    21 The usual Deligne conjecture (proved variously in \cite{MR1805894, MR2064592, hep-th/9403055, MR1805923} gives a map
    18 The usual Deligne conjecture (proved variously in \cite{MR1805894, MR2064592, hep-th/9403055, MR1805923} gives a map
    22 \[
    19 \[
    23 	C_*(LD_k)\otimes \overbrace{Hoch^*(C, C)\otimes\cdots\otimes Hoch^*(C, C)}^{\text{$k$ copies}}
    20 	C_*(LD_k)\otimes \overbrace{Hoch^*(C, C)\otimes\cdots\otimes Hoch^*(C, C)}^{\text{$k$ copies}}
    24 			\to  Hoch^*(C, C) .
    21 			\to  Hoch^*(C, C) .
    25 \]
    22 \]
    26 Here $LD_k$ is the $k$-th space of the little disks operad, and $Hoch^*(C, C)$ denotes Hochschild
    23 Here $LD_k$ is the $k$-th space of the little disks operad and $Hoch^*(C, C)$ denotes Hochschild
    27 cochains.
    24 cochains.
    28 The little disks operad is homotopy equivalent to the 
    25 The little disks operad is homotopy equivalent to the 
    29 (transversely orient) fat graph operad
    26 (transversely oriented) fat graph operad
    30 \nn{need ref, or say more precisely what we mean}, 
    27 (see below), 
    31 and Hochschild cochains are homotopy equivalent to $A_\infty$ endomorphisms
    28 and Hochschild cochains are homotopy equivalent to $A_\infty$ endomorphisms
    32 of the blob complex of the interval, thought of as a bimodule for itself.
    29 of the blob complex of the interval, thought of as a bimodule for itself.
    33 \nn{need to make sure we prove this above}.
    30 (see \S\ref{ss:module-morphisms}).
    34 So the 1-dimensional Deligne conjecture can be restated as
    31 So the 1-dimensional Deligne conjecture can be restated as
    35 \[
    32 \[
    36 	C_*(FG_k)\otimes \hom(\bc^C_*(I), \bc^C_*(I))\otimes\cdots
    33 	C_*(FG_k)\otimes \hom(\bc^C_*(I), \bc^C_*(I))\otimes\cdots
    37 	\otimes \hom(\bc^C_*(I), \bc^C_*(I))
    34 	\otimes \hom(\bc^C_*(I), \bc^C_*(I))
    38 	  \to  \hom(\bc^C_*(I), \bc^C_*(I)) .
    35 	  \to  \hom(\bc^C_*(I), \bc^C_*(I)) .
    71 \caption{An $n$-dimensional fat graph}\label{delfig2}
    68 \caption{An $n$-dimensional fat graph}\label{delfig2}
    72 \end{figure}
    69 \end{figure}
    73 
    70 
    74 More specifically, an $n$-dimensional fat graph ($n$-FG for short) consists of:
    71 More specifically, an $n$-dimensional fat graph ($n$-FG for short) consists of:
    75 \begin{itemize}
    72 \begin{itemize}
    76 \item ``Upper" $n$-manifolds $M_0,\ldots,M_k$ and ``lower" $n$-manifolds $N_0,\ldots,N_k$,
    73 \item ``Lower" $n$-manifolds $M_0,\ldots,M_k$ and ``upper" $n$-manifolds $N_0,\ldots,N_k$,
    77 with $\bd M_i = \bd N_i = E_i$ for all $i$.
    74 with $\bd M_i = \bd N_i = E_i$ for all $i$.
    78 We call $M_0$ and $N_0$ the outer boundary and the remaining $M_i$'s and $N_i$'s the inner
    75 We call $M_0$ and $N_0$ the outer boundary and the remaining $M_i$'s and $N_i$'s the inner
    79 boundaries.
    76 boundaries.
    80 \item Additional manifolds $R_1,\ldots,R_{k}$, with $\bd R_i = E_0\cup \bd M_i = E_0\cup \bd N_i$.
    77 \item Additional manifolds $R_1,\ldots,R_{k}$, with $\bd R_i = E_0\cup \bd M_i = E_0\cup \bd N_i$.
    81 %(By convention, $M_i = N_i = \emptyset$ if $i <1$ or $i>k$.)
    78 %(By convention, $M_i = N_i = \emptyset$ if $i <1$ or $i>k$.)
   108 leaving the $M_i$ and $N_i$ fixed.
   105 leaving the $M_i$ and $N_i$ fixed.
   109 (Keep in mind the case $R'_i = R_i$.)
   106 (Keep in mind the case $R'_i = R_i$.)
   110 (See Figure \ref{xdfig3}.)
   107 (See Figure \ref{xdfig3}.)
   111 \begin{figure}[t]
   108 \begin{figure}[t]
   112 $$\mathfig{.4}{deligne/dfig3a} \to \mathfig{.4}{deligne/dfig3b} $$
   109 $$\mathfig{.4}{deligne/dfig3a} \to \mathfig{.4}{deligne/dfig3b} $$
   113 \caption{Conjugating by a homeomorphism}\label{xdfig3}
   110 \caption{Conjugating by a homeomorphism
       
   111 \nn{change right $R_i$ to $R'_i$}}\label{xdfig3}
   114 \end{figure}
   112 \end{figure}
   115 \item If $M_i = M'_i \du M''_i$ and $N_i = N'_i \du N''_i$ (and there is a
   113 \item If $M_i = M'_i \du M''_i$ and $N_i = N'_i \du N''_i$ (and there is a
   116 compatible disjoint union of $\bd M = \bd N$), we can replace
   114 compatible disjoint union of $\bd M = \bd N$), we can replace
   117 \begin{eqnarray*}
   115 \begin{eqnarray*}
   118 	(\ldots, M_{i-1}, M_i, M_{i+1}, \ldots) &\to& (\ldots, M_{i-1}, M'_i, M''_i, M_{i+1}, \ldots) \\
   116 	(\ldots, M_{i-1}, M_i, M_{i+1}, \ldots) &\to& (\ldots, M_{i-1}, M'_i, M''_i, M_{i+1}, \ldots) \\
   168 Let $f_i = \rm{id}$ for all $i$.
   166 Let $f_i = \rm{id}$ for all $i$.
   169 We have now defined a map from the little $n{+}1$-ball operad to the $n$-FG operad,
   167 We have now defined a map from the little $n{+}1$-ball operad to the $n$-FG operad,
   170 with contractible fibers.
   168 with contractible fibers.
   171 (The fibers correspond to moving the $D_i$'s in the $x_{n+1}$ 
   169 (The fibers correspond to moving the $D_i$'s in the $x_{n+1}$ 
   172 direction without changing their ordering.)
   170 direction without changing their ordering.)
   173 \nn{issue: we've described this by varying the $R_i$'s, but above we emphasize varying the $f_i$'s.
   171 %\nn{issue: we've described this by varying the $R_i$'s, but above we emphasize varying the $f_i$'s.
   174 does this need more explanation?}
   172 %does this need more explanation?}
   175 
   173 
   176 Another familiar subspace of the $n$-FG operad is $\Homeo(M\to N)$, which corresponds to 
   174 Another familiar subspace of the $n$-FG operad is $\Homeo(M_0\to N_0)$, which corresponds to 
   177 case $k=0$ (no holes).
   175 case $k=0$ (no holes).
   178 
   176 
   179 \medskip
   177 \medskip
   180 
   178 
   181 Let $\ol{f} \in FG^n_{\ol{M}\ol{N}}$.
   179 Let $\ol{f} \in FG^n_{\ol{M}\ol{N}}$.
   192 				 \stackrel{\id\ot\alpha_1}{\to} \bc_*(R_1\cup N_1)
   190 				 \stackrel{\id\ot\alpha_1}{\to} \bc_*(R_1\cup N_1)
   193 				 \stackrel{f_1}{\to} \bc_*(R_2\cup M_2) \stackrel{\id\ot\alpha_2}{\to}
   191 				 \stackrel{f_1}{\to} \bc_*(R_2\cup M_2) \stackrel{\id\ot\alpha_2}{\to}
   194 				 \cdots  \stackrel{\id\ot\alpha_k}{\to} \bc_*(R_k\cup N_k)
   192 				 \cdots  \stackrel{\id\ot\alpha_k}{\to} \bc_*(R_k\cup N_k)
   195 				 \stackrel{f_k}{\to} \bc_*(N_0)
   193 				 \stackrel{f_k}{\to} \bc_*(N_0)
   196 \]
   194 \]
   197 (Recall that the maps $\id\ot\alpha_i$ were defined in \nn{need ref}.)
   195 (Recall that the maps $\id\ot\alpha_i$ were defined in \S\ref{ss:module-morphisms}s.)
   198 \nn{need to double check case where $\alpha_i$'s are not closed.}
   196 \nn{need to double check case where $\alpha_i$'s are not closed.}
   199 It is easy to check that the above definition is compatible with the equivalence relations
   197 It is easy to check that the above definition is compatible with the equivalence relations
   200 and also the operad structure.
   198 and also the operad structure.
   201 We can reinterpret the above as a chain map
   199 We can reinterpret the above as a chain map
   202 \[
   200 \[
   235 \]
   233 \]
   236 It suffices to show that the above maps are compatible with the relations whereby
   234 It suffices to show that the above maps are compatible with the relations whereby
   237 $FG^n_{\overline{M}, \overline{N}}$ is constructed from the various $P$'s.
   235 $FG^n_{\overline{M}, \overline{N}}$ is constructed from the various $P$'s.
   238 This in turn follows easily from the fact that
   236 This in turn follows easily from the fact that
   239 the actions of $C_*(\Homeo(\cdot\to\cdot))$ are local (compatible with gluing) and associative.
   237 the actions of $C_*(\Homeo(\cdot\to\cdot))$ are local (compatible with gluing) and associative.
   240 
   238 %\nn{should add some detail to above}
   241 \nn{should add some detail to above}
       
   242 \end{proof}
   239 \end{proof}
   243 
   240 
   244 \nn{maybe point out that even for $n=1$ there's something new here.}
   241 We note that even when $n=1$, the above theorem goes beyond an action of the little disks operad.
       
   242 $M_i$ could be a disjoint union of intervals, and $N_i$ could connect the end points of the intervals
       
   243 in a different pattern from $M_i$.
       
   244 The genus of the fat graph could be greater than zero.