Professional Documents
Culture Documents
Textbook Geometric and Topological Aspects of The Representation Theory of Finite Groups Pims Summer School and Workshop July 27 August 5 2016 1St Edition Jon F Carlson Ebook All Chapter PDF
Textbook Geometric and Topological Aspects of The Representation Theory of Finite Groups Pims Summer School and Workshop July 27 August 5 2016 1St Edition Jon F Carlson Ebook All Chapter PDF
https://textbookfull.com/product/geometric-methods-in-physics-
xxxv-workshop-and-summer-school-bialowieza-poland-
june-26-july-2-2016-1st-edition-piotr-kielanowski/
https://textbookfull.com/product/topological-aspects-of-
condensed-matter-physics-lecture-notes-of-the-les-houches-summer-
school-volume-103-august-2014-1st-edition-claudio-chamon/
https://textbookfull.com/product/representation-theory-of-finite-
groups-a-guidebook-1st-edition-david-a-craven/
https://textbookfull.com/product/representation-theory-of-finite-
groups-a-guidebook-1st-edition-david-a-craven-2/
Geometric Methods in Physics XXXVII Workshop and Summer
School Bia■owie■a Poland 2018 Piotr Kielanowski
https://textbookfull.com/product/geometric-methods-in-physics-
xxxvii-workshop-and-summer-school-bialowieza-poland-2018-piotr-
kielanowski/
https://textbookfull.com/product/theory-of-groups-and-symmetries-
finite-groups-lie-groups-and-lie-algebras-1st-edition-alexey-p-
isaev/
https://textbookfull.com/product/representation-theory-of-finite-
monoids-1st-edition-benjamin-steinberg/
https://textbookfull.com/product/a-journey-through-
representation-theory-from-finite-groups-to-quivers-via-algebras-
caroline-gruson/
https://textbookfull.com/product/complex-semisimple-quantum-
groups-and-representation-theory-1st-edition-christian-voigt/
Springer Proceedings in Mathematics & Statistics
Jon F. Carlson
Srikanth B. Iyengar
Julia Pevtsova Editors
Geometric and
Topological Aspects
of the Representation
Theory of Finite
Groups
PIMS Summer School and Workshop,
July 27–August 5, 2016
Springer Proceedings in Mathematics & Statistics
Volume 242
Springer Proceedings in Mathematics & Statistics
This book series features volumes composed of selected contributions from
workshops and conferences in all areas of current research in mathematics and
statistics, including operation research and optimization. In addition to an overall
evaluation of the interest, scientific quality, and timeliness of each proposal at the
hands of the publisher, individual contributions are all refereed to the high quality
standards of leading journals in the field. Thus, this series provides the research
community with well-edited, authoritative reports on developments in the most
exciting areas of mathematical and statistical research today.
Julia Pevtsova
Editors
123
Editors
Jon F. Carlson Julia Pevtsova
Department of Mathematics Department of Mathematics
University of Georgia University of Washington
Athens, GA, USA Seattle, WA, USA
Srikanth B. Iyengar
Department of Mathematics
University of Utah
Salt Lake City, UT, USA
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
This volume is a proceedings of two workshops that were held in the summers of
2015 and 2016. Both events celebrated the mathematical accomplishments of
David J. Benson for the occasion of his 60th birthday. The papers presented at the
meetings were representative of the many mathematical subjects that Dave has
worked on during his distinguished career. The extent of his influence in these
areas, which was clearly on display at both of the meetings, is partly reflected in the
fact that there were two meetings, within the space of a year, organized in Dave’s
honor!
Dave did his graduate study at Cambridge University and wrote his thesis under
the direction of John Thompson. He is well known for establishing new relation-
ships between diverse mathematical areas, particularly connecting group repre-
sentation theory with commutative algebra and homotopy theory. In addition to his
many papers, his conjectures on subjects ranging from branching rules for modules
over symmetric groups to the regularity of cohomology rings of finite groups have
generated much activity. His books, most notably the two-volume survey of group
representations and cohomology, have been standard references for years and an
inspiration for many of the advances in the subject. Dave’s impact on the field has
also been through his numerous collaborations—he has 50 collaborators, according
v
vi Preface
to Math. Sci. Net., including the three editors of this volume. Dave is also much
admired by his colleagues and friends not only for being a fantastic mathematician
but for the whole string of other talents and interests he displays, including juggling,
unicycling, opera singing, and writing a popular book on “Math and Music”.
The first workshop had the title “Groups, Representations and Cohomology” and
was held on June 22–27, 2015 at Sabhal Mòr Ostaig, a school on the Isle of Skye in
Scotland. The organizers were Radha Kessar, Henning Krause, Markus Linckelmann,
Ran Levi, and Peter Symonds. It was preceded by an Anglo-Franco-German
Representation Theory Network Summer School on the subject “Support Varieties in
Modular Representation Theory” held on June 18–22, 2015, organized by Ivo
Dell’Ambrogio, Greg Stevenson, and Jan Stovicek.
The second meeting consisted of a summer school and workshop on the subject
“Geometric Methods in the Representation Theory of Finite Groups”. It took place
at the Pacific Institute for the Mathematical Sciences at the University of British
Columbia in Vancouver, from July 27 to August 5 of 2016. The summer school
featured four series of lectures by Eric Friedlander, Jesper Grodal, Radha Kessar,
and Peter Symonds. The summer school and workshop were organized by the three
editors of these proceedings.
Of the contributors to this volume, Carlson, Erdmann, Greenlees, Iyengar,
Oliver, Pevtsova, and Touzé presented invited lectures at the meeting on the Isle of
Skye. Friedlander, Kessar, and Symonds gave series of several lectures each at the
summer school at PIMS. Their submissions included here are summaries of those
lectures. Avramov, Henke, Huisgen-Zimmermann, Kessar, Krause, Linckelmann,
Lynd, Mathew, Nakano, Sobaje, Stevenson, Webb, and Witherspoon were invited
to speak at the conference in Vancouver.
Partial funding for the participants and speakers at the Vancouver meeting was
also provided by the National Security Agency (under grant NSA-AMS-141009)
and the National Science Foundation (under grant DMS-1624050). The meeting
was also supported by funds from the Sondersforschungsbereich 701 at Bielefeld
Universität. It remains to thank the Pacific Institute for the Mathematical Sciences
for their help in organizing the meeting in Vancouver and this volume.
vii
viii Contents
Partly supported by NSF grants DMS-1103176 (LLA) and DMS-1700985 (SBI). We are grateful
to Chris Drupieski for comments on earlier version of this manuscript.
L. L. Avramov
Department of Mathematics, University of Nebraska, Lincoln, NE 68588, U.S.A.
e-mail: avramov@math.unl.edu
S. B. Iyengar (B)
Department of Mathematics, University of Utah, Salt Lake City, UT 84112, U.S.A.
e-mail: iyengar@math.utah.edu
(k, M ) ∼
P/(f ) P/(g)
Tor i = Tor i (k, M ) for each i ≥ 0.
In this section, we recall what little is needed by way of constructs and results
concerning DG algebras and DG modules. Our reference for this material is [3].
Let P be a commutative ring and A a DG P-algebra; it will be implicit that A is
graded commutative and satisfies Ai = 0 for i < 0.
(a ⊗ b)(a ⊗ b ) = (−1)|a ||b| aa ⊗ bb .
For w ∈ A2d with d ≥ 1, we say that (w (i) ∈ A2di )i0 is a sequence of divided powers
of w if w (0) = 1, w (1) = w, and there are equalities
(i + j)! (i+j)
w (i) w (j) = w and ∂(w (i) ) = ∂(w)w (i−1) for all i, j ≥ 0 .
i!j!
The right-hand side is well defined because the expression of w is unique. A direct
computations shows that ∂(w(i) ) = ∂(w)w (i−1) holds for i ≥ 0.
4 L. L. Avramov and S. B. Iyengar
Let P y be a graded-free P-module with basis {y(i) }i0 , where |y(i) | = 2i, and define
a P-linear multiplication table by
(i + j)! (i+j)
y(0) = 1 and y(i) y(j) = y for all i, j ≥ 0 .
i!j!
Hi (A) −→ Hi (A y )
that is bijective for i = 0 and surjective for i = 1, with kernel the ideal generated by
the class of z. The results below and its proof are standard; see [3, Sect. 6]. Details
of the proof are given for ease of reference.
Lemma 1.1 Let z be a cycle in A1 .
(1) If an element w ∈ A2 has a sequence of divided powers, then the DG P-algebras
A y | ∂(y) = z and A y | ∂(y) = z + ∂(w) are isomorphic.
(2) If the class of z is a basis for the H0 (A)-module H1 (A) and Hi (A) = 0 for i ≥ 2,
then the canonical map
A y | ∂(y) = z −→ H0 (A)
is a quasi-isomorphism.
Proof It is readily verified that the A-linear map
0 −→ A −→ A y −→ 2 A y −→ 0 .
ð
0 −→ H2 (A y ) −→ H0 (A y ) −→ H1 (A) −→ H1 (A y ) −→ 0
Hi (A y ) ∼
= Hi−2 (A y ) for i ≥ 3.
Let t be a finite set of elements generating an ideal n of P and K the Koszul complex
on t. Let z1 , . . . , zc be cycles in K1 that generate the H0 (K)-module H1 (K). Iterating
the construction from Sect. 1.4 yields a DG P-algebra
T := K y1 , . . . , yc | ∂(yi ) = zi .
2 Hypersurfaces
πf : P/(f ) −→ R
Theorem 2.1 Let I ⊂ P be an ideal containing a regular element, set R := P/I and
let M be an R-complex.
(1) For each f ∈ I , there exists an h ∈ nI such that f + h is P-regular.
(2) Let f and g be regular elements lying in I . If f − g is in nI , then there is an
isomorphism of graded k-vector spaces
The proof is given at the end of this section. Part (1) is well known and part (3) is
essentially contained in the work of Shamash [14, Theorem 1].
Throughout the rest of the section, t := t1 , . . . , tn denotes a finite generating set
for the ideal n and A denotes an acyclic closure of k over P; see Sect. 1.5.
H(C) ∼
= Tor P (k, R) .
Z := {(f , x) ∈ I ⊕ A1 | f = ∂(x)} .
This is a P-submodule of I ⊕ A1 .
Lemma 2.2 The assignments f ← (f , x) → x ⊗ 1 define surjective P-linear maps
ε ζ
I←− Z −→ Z1 (C) .
Conversely,
f in nI gives f = ∂( ai xi ) with {xi }ni=1 in A
1 and {ai }i=1 in I , so that
n
2.2 Hypersurfaces
Bf := A y | ∂(y) = z f .
Proof Applying Lemma 2.2 with (f ) in place of I , we see that H1 (A) is generated
by the class of zf and is zero if and only if f lies in nf ; the latter possibility is ruled
out by the hypothesis that f is regular. By using the resolution
f
0→P−
→ P → 0,
of P over P we get
k for i = 0, 1
Tor Pi (k, P) ∼
=
0 otherwise
Since H(A) ∼
= Tor P (k, P), it remains to apply Lemma 1.1.
Proof of Theorem 2.1 The zero divisors in P are the elements of its associated primes;
call them p1 , . . . , pn . Since I has a regular element so does nI , and hence the latter
is not contained in ∪ni=1 pi . Thus, (1) follows from [13, Theorem 124].
In the next steps, the notation and constructions from Sects. 2.1 and 2.2 will be
used. In particular, A is a DG P-algebra resolution of k, and C := A ⊗P R. Moreover
Bf is a DG P-algebra resolution of k.
C y | ∂(y) = zf ⊗ 1 ⊗R M
Indeed, Tor P/(f ) (k, M ) is the homology of the DG R-module Bf ⊗P/(f ) M . Asso-
ciativity of tensor products yields isomorphisms of R-complexes
8 L. L. Avramov and S. B. Iyengar
Bf ⊗P/(f ) M ∼
= ((A/fA) y | ∂(y) = z f ⊗P/(f ) R) ⊗R M
∼
= C y | ∂(y) = zf ⊗ 1 ⊗R M
(3) When f ∈ nI holds, the element zf is in IA and hence its residue class in C is
zero. The preceding claim then yields
C y | ∂(y) = zf ⊗ 1 ∼
= C y | ∂(y) = zg ⊗ 1 .
Indeed, Lemma 2.2 provides elements (f , x) and (g, y) in the module Z defined
in Sect. 2.1, and w in C2 satisfying x ⊗ 1 − y ⊗ 1 = ∂(w). Since w has a sequence
of divided powers (see Sect. 1.5), Lemma 1.1 yields the desired isomorphism.
The isomorphism in the Claim 2 induces an isomorphism of R-complexes
C y | ∂(y) = zf ⊗ 1 ⊗R M ∼
= C y | ∂(y) = zg ⊗ 1 ⊗R M .
3 Support Sets
Let (P, n, k) be a (commutative noetherian) local ring, with maximal ideal n and
residue field k. Let π : P → R be a surjective homomorphism of rings such that the
ideal I := Ker(π) contains a regular element. Throughout, M will be an R-module;
the results presented below all carry over to complexes; see Remark 3.5.
We view the k-vector space I /nI as endowed with the Zariski topology. Given f ∈ I
we write [f ] for its residue class in I /nI . The homological support set of M with
Restricting Homology to Hypersurfaces 9
We write proj dimR M for the projective dimension of the R-module M . Assume M
is finitely generated or m-torsion; the latter condition means that for each u ∈ M
there is an integer l such that ml u = 0. Then
Fix a minimal generating set f1 , . . . , fc for the ideal I . The residue classes [f1 ], . . . , [fc ]
form a basis for I /nI as a k-vector space. We will use this to identify I /nI with k c
whenever necessary, or convenient.
Given a point a := (a1 , . . . , ac ) in k c , it follows from Theorem 2.1(1) that there
is an element a := (a1 , . . . , ac ) in P c such that
i = ai mod n for i = 1, . . . , c;
(1) a
(2) ai fi is regular in P.
We will call such an element a a lifting of a to P. One can then also describe the
subset Vπ (M ) from Sect. 3.1 as
10 L. L. Avramov and S. B. Iyengar
there exists a lifting a of a with
Vπ (M ) = a ∈ k c
proj dimP/(f ) (M ) = ∞ for f = ai fi
In particular, the right-hand side does not depend on the choice of a minimal gener-
ating set for I . The subset Vπ (M ) ⊆ k c is closed when M is finitely generated; this is
contained in Theorem 3.4. For the proof of this result, we require a statement about
complexes over regular rings, presented below.
By a regular ring we mean a commutative noetherian ring S such that the local rings
Sp are regular for all p in Spec S. For any p ∈ Spec S, We write k(p) for the residue
field, (S/p)p , at p.
Lemma 3.1 Let S be a regular ring and let X be a complex of S-modules such that
Hi (X ) is finitely generated for each i and equals 0 for i 0. Fix a prime ideal p in
S. The following conditions are equivalent.
(1) Tor Si (k(p), X ) = 0 for i 0;
(2) k(p) ⊗S Hi (X ) = 0 for i 0;
(3) Hi (X )p = 0 for i 0.
Proof Since the action S on k(p) factors through the localization homomorphism
S → Sp and Sp is flat as an S-module, one has isomorphisms
Tor S (k(p), X ) ∼
= Tor Sp (k(p), Xp )
k(p) ⊗S Hi (X ) ∼ = k(p) ⊗S Hi (X )p
p
c
:= P[s]/(
Q f) where
f := s i fi .
i=1
f
0 −→ P[s] −→ P[s] −→ 0 (2)
is a free resolution of Q.
For each a := (a1 , . . . , ac ) in P c form, the homomorphism of P-algebras
In what follows, we write Pa for P viewed as a P[s]-module via εa . Base change
gives a commutative square
along the canonical surjection P[s] → Q
εa
P[s] P
Q Q P/(f )
where f = ai fi .
Lemma 3.2 When the element f := ai fi is P-regular, one has
Q for i = 0
Tor P[s]
i (Pa , Q) =
0 otherwise
f
0 −→ P −→ P −→ 0
Set M [s] := P[s] ⊗P M . Evidently f · M [s] = 0 so M [s] is an Q-module.
(2) When i ai fi is regular in P, one has
Q M for i = 0
Tor i (Q, M [s]) =
0 otherwise
Proof The argument uses the standard first quadrant change of rings spectral
and a P[s]-module L.
sequence associated to the homomorphism P[s] → Q,
2
Ei,j
Q
:= Tor i (Tor P[s] P[s]
j (L, Q), M [s]) =⇒ Tor i+j (L, M [s])
2
Thus, the Ei,j = 0 for j = 0, 1, so the spectral sequence unwinds to an exact sequence
Q
Q
· · · → Tor P[s]
i+2 (N , M [s]) → Tor i+2 (N , M [s]) → Tor i (N , M [s])
→ Tor P[s]
i+1 (N , M [s]) → · · ·
Since proj dimP M = p, one has proj dimP[s] M [s] = p, so the first and the last
modules in the sequence above are zero for i ≥ p. This yields (1).
(2) Consider the spectral sequence above with L := Pa . In view of Lemma 3.2,
this spectral sequence collapses on the second page and yields isomorphisms
Tor i (Q, M [s]) ∼
Q P[s]
= Tor i (Pa , M [s]) for each i.
The statement and proof of the next result involves constructions introduced in
Sect. 3.5. We identify k[s] and the ring of k-valued algebraic functions on I /nI by
mapping si to the ith coordinate function of the k-basis {[f1 ], · · · , [fc ]} of of I /nI .
Restricting Homology to Hypersurfaces 13
Theorem 3.4 Assume that k is algebraically closed and that I contains a regular
element. Fix an integer d ≥ depth P. Let M be a finitely generated R-module and J
the annihilator of the k[s]-module
Q
Q
Tor d (k[s], M [s]) ⊕ Tor d +1 (k[s], M [s]) .
Proof Set p := proj dimP M . If proj dimP/(f ) M is finite for some regular element
f ∈ I , then p is finite as well. Therefore, Vπ (M ) = k c when p = ∞; see Sect. 3.3.
For the rest of the proof, we assume that p is finite.
a ∈ Ak . Let a ∈ P be a lifting of a, as in Sect. 3.3, and set Q := P/(f ),
c c
Fix a point
where f = i ai fi . Let G be a free resolution of M [s] over Q and set
G.
X := k[s] ⊗Q
In particular, the k[s]-module Hi (X ) is finitely generated for each i, and 0 for i < 0.
Tor Q (k, M ) ∼
= Tor k[s] (ka , X ) .
k ⊗Q (Q ⊗Q ∼
G) = k ⊗Q
∼
G = ka ⊗k[s] (k[s] ⊗Q
G) = ka ⊗k[s] X .
a ∈ Vπ (M ) ⇐⇒ ma ⊇ J .
14 L. L. Avramov and S. B. Iyengar
Indeed, consider the following chain of equivalences, where the first one holds by
definition, and the second is by Claim 1.
Q
/ Vπ (M ) ⇐⇒ Tor i (k, M ) = 0 for i 0
a∈
⇐⇒ Tor k[s]
i (ka , X ) = 0 for i 0
⇐⇒ Hi (X )ma = 0 for i 0.
The last one is given by Lemma 3.1, applied with S := k[s] and p := ma . Since
proj dimR M = p is finite, the Auslander–Buchsbaum Equality implies d ≥ p. Propo-
sition 3.3(1), then gives the first equivalence below.
The second one is standard. Putting together the string of equivalences above yields
the contrapositive of the desired claim.
Combining this with the conclusion in Claim 2 yields the desired result.
Remark 3.5 Theorem 3.4 carries over, with essentially the same proof, to the case
when M is an R-complex with H(M ) finitely generated. The crucial point is that, in
the notation above, the natural morphism of Q-complexes
M [s] −→ M
Q ⊗LQ
Remark 3.6 When P is a domain the homological support set of M , as in Sect. 3.1, is
the one introduced by Jorgensen [12], which extends the notion of varieties for mod-
ules over complete intersections due to Avramov [2]. The last part of Theorem 3.4—
that Vπ (M ) is closed—was proved by Jorgensen [12, Theorem 2.2] under the addi-
tional assumption that P is a domain. The argument in op. cit. is quite different from
the one presented above, which builds on a idea in the proof of [4, Theorem 3.1].
Restricting Homology to Hypersurfaces 15
4 Defining Equations
The main result of this section, Theorem 4.3, gives equations that define the homo-
logical support set of modules over complete intersections. The statement, and its
proof, involves some linear algebra over commutative rings, recalled below.
where Ir is the identity matrix of size r; see also [7, Lemmas 1.4.8, 1.4.9].
√
The results below, where J denotes the radical of an ideal J , concerns differential
modules; cf. [5, Remark 1.6], and also [5, Example 1.7] that shows that the hypothesis
that S is regular is needed.
Proof Set H := Ker(δ)/ Im(δ). The radical of an ideal is the intersection of the
prime ideals containing it, and a prime ideal p of S contains J precisely when Hp is
non-zero. Thus, the desired statement is equivalent to:
Hp = 0 ⇐⇒ Ir (δ) ⊂ p .
0 −→ Im(δ) −→ S 2r −→ · · · −→ S 2r −→ Im(δ) −→ 0 ,
of length n. In particular for n = dim S + 2, one sees that Im(δ) is a (dim S)-th
syzygy module and hence free, as S is a regular local ring. This is as desired.
Assume Ir (δ) = S. Since S is a domain and rank S Im(δ) ≤ r, it follows from
Remark 4.1 that Coker(δ) is free of rank r. This implies that, with l the residue field
of S, the natural maps
∼
= ∼
=
Coker(δ) ⊗S l −→ Coker(δ ⊗S l) and Im(δ) ⊗S l −→ Im(δ ⊗S l)
are isomorphisms. Thus, applying (−) ⊗S l to (5) and keeping in mind that Im(δ) is
free, one then gets an exact sequence
0 → H ⊗S l → Coker(δ ⊗S l) → Im(δ ⊗S l) → 0 ,
The isomorphisms above imply that both l-vector spaces on the right have rank r. It
follows that H ⊗S l = 0; thus H = 0, by Nakayama’s Lemma.
Let (P, n, k) be a regular local ring of Krull dimension d and containing its residue
field, k, as a subring. Let I be the ideal generated by a P-regular sequence f1 , . . . , fc
in n2 and set R := P/I .
As in Sect. 3.5, let P[s] be the polynomial ring over P in indeterminates s :=
s1 , . . . , sc , set := P[s]/(
f := i si fi and Q is a hypersurface, the Q-
f ). Since Q
Restricting Homology to Hypersurfaces 17
Q
module Q/n ∼ = k[s] has a free resolution G with the property that G i ∼
=Q2d −1 and
∂i = ∂i+2 for all i ≥ c so that the complex G ≥c has the form
B A B A
· · · −→ G c+3 −→ G c+2 −→ G c+1 −→ G c
When M has finite rank as a k-vector space, one can view this as an endomorphism
of free k[s]-modules of rank (rank k M )2d ; we write C(M ) for this map.
d C(M ) d C(M ) d
M [s]2 −−−−→ M [s]2 −−−−→ M [s]2
Remark 4.4 Theorem 4.3 should be compared to [4, Theorem 3.2] that describes
equations defining VR (M ) in terms of data derived from a free resolution, say F, of
M over P, instead of the resolution of k[s] over Q. In detail, one can construct a
system of higher homotopies, a family of endomorphism (not chain maps) of F that
encode the data that, because M is an R-module, multiplication by f1 , . . . , fc on F is a
commuting family of morphisms that are homotopic to zero. Then [4, Theorem 3.2],
expresses Vπ (M ) as the zero set of appropriate minors of the matrix involving these
higher homotopies. In contrast, the C(M ) appearing in Theorem 4.3 is universal, in
that it involves only a matrix factorization of
f , and matrices describing the action
of the elements of P on M , viewed as k-vector space.
We finish by describing the free resolution of k[s] over Q and, in particular, the
matrices A and B, when R is a truncated power series ring; group algebras of abelian
p-groups over a field of positive characteristic p have this form.
18 L. L. Avramov and S. B. Iyengar
Example 4.5 Let k be a field, P := k[|t1 , . . . , td |] the power series ring over k in
indeterminates t := t1 , . . . , td , and set
is a Q-free
resolution of Q/(t), that is to say, of k[s]; see Lemma 2.2. For any integer
n ≥ c, one has
E2i when n is even
E y n := i0
i0 E2i+1 when n is odd
In either parity, it is a finite free Q-module of rank 2c−1 . Moreover, after a suitable
choice of bases, the differential on E y is given by
described below.
with A, B are square matrices of size 2c−1 , and coefficients in Q,
The rows and columns of A and B are indexed by subsets of the sequences
πup : P/(up ) −→ R
In the equivalence above, the invariant on the right is the flat dimension.
Let Q be a commutative noetherian ring and N a Q-module. Recall that flat dimQ (N ),
the flat dimension of N , is the length of the shortest flat resolution of N over Q. Evi-
dently flat dimQ (N ) ≤ proj dimQ (N ); equality holds when N is finitely generated.
The Q-module N is J -torsion, for some ideal J in Q, if for each u ∈ N there is an
integer l such that J l u = 0. If N is n-torsion for a maximal ideal n of Q, then
Q
flat dimQ N < ∞ ⇐⇒ Tor i (Q/n, N ) = 0 fori 0 . (6)
This follows form [6, Propositions 5.3.F], keeping in mind that for any prime ideal
q of Q, when q = n, one has Tor Q ((Q/q)q , Nq ) = 0.
We record a few consequences of Theorem 5.1.
Corollary 5.2 If an element w ∈ n satisfies u − w ∈ n2 , then M ↓w is free if and
only if is M ↓u .
Proof Evidently nl annihilates R, and hence M , for l 0; thus M is n-torsion. Given
(6), the desired result follows from Theorems 2.1(2) and 5.1.
20 L. L. Avramov and S. B. Iyengar
In this paragraph we assume that k is algebraically closed. Let Ack be the affine space
over k, of dimension c. We write a for a point (a1 , . . . , ac ) in this space. For each
R-module M set
VRr (M ) := a ∈ Ack M ↓u is not free for u = ai si ∪ {0} .
i
This is called the rank variety of M . It was introduced by Carlson who also proved
that it is a closed subset of Ack ; see [8, §4]. The result below, which relates the rank
variety of M to its homological support set in the sense of Sect. 3.1, is contained
in [2, Theorem 7.5]. The argument in op. cit. gets to the stated bijection via the
cohomological variety of M . Our proof, using Theorems 2.1 and 5.1, is more direct.
p p
Corollary 5.3 The map Ack → Ack sending (a1 , . . . , ac ) to (a1 , . . . , ac ) is bijective
and for each R-module M it restricts to a bijection
∼
=
VRr (M ) −→ VR (M ) .
Proof The given map is a bijection on Ack because k is algebraically closed. The
assertion about the varieties associated to M is a translation of Theorem 5.1.
Proof The inequality on the left holds because any flat resolution over S[x] is also a
flat resolution over S, due to the freeness of S[x] as a S-module.
For the inequality on the right, we may assume that flat dimS M is finite. Induction
on d then shows that it suffices to deal with the case when S[x] is a polynomial ring in
a single indeterminate. In the ring S[x] ⊗S S[x], the element x ⊗ 1 − 1 ⊗ x is regular,
so we have an exact sequence of S[x] ⊗S S[x]-modules
x⊗1−1⊗x
0 −→ (S[x] ⊗S S[x]) −−−−−−→ (S[x] ⊗S S[x]) −→ S[x] → 0
Another random document with
no related content on Scribd:
T
Taffy’s imagination (fear of the dark)
MP25627.
Taj Mahal.
MP26015.
The Take.
LP43762.
Take my life.
R593268.
The Taking of Pelham one two three.
LP43957.
Tale of King Midas.
MP25885.
Tale of Rumpelstiltskin.
MP25905.
Tales of Manhattan.
LU3678.
The Talking magpies in Flying south.
R591491.
Tanbark champions.
R583809.
Tandem Productions, Inc.
LP43835 - LP43853.
LP43965 - LP43981.
Tantalus, Inc.
MP25507.
Tawny Pipit.
R593272.
Tax audit.
LP43849.
Taylor, Stanley B.
MP25706.
The Teachers’ crisis.
MP26037.
Teaching Films, Inc., division of A-V Corporation. SEE A-V
Corporation. Teaching Films, Inc.
Technique for lifting.
MP25852.
Ted Ziegler and his folding dome meet Bucky Fuller.
MU9077.
Teenage relationships: Venessa and her friends.
MP25908.
Telephone techniques to get appointments.
MP25990.
Tell ’em where to get it.
MP26055.
Tenafly.
LP43708.
Tender years.
R589461.
Tennis in rhythm.
R584970.
Tennis town.
R590801.
Tenth Avenue angel.
R594212.
Terminal man.
LP43787.
Terr-Filmkunst.
R584014.
Territoriality in reef animals.
MP25866.
Terry-Toons, Inc.
R591488 - R591494.
The Texas chain saw massacre.
LU3676.
Texas. Criminal Justice Division.
MP25813 - MP25814.
MP25993 - MP25996.
Text of light.
MP25816.
Texture Films, Inc.
LP43674.
LP43675.
The T-group experience for individual and organizational
effectiveness, film 1.
MP25609.
The T-group experience for individual and organizational
effectiveness, film 2.
MP25610.
The T-group experience for individual and organizational
effectiveness, film 3.
MP25611.
Thailand: past and present.
MP25705.
That Hagen girl.
R592812.
That’s entertainment.
LP43640.
That’s my gal.
R582730.
That’s my man.
R582729.
They’re off.
R580998.
They steal cars, don’t they?
LP43697.
They won’t believe me.
R584159.
Thicker.
MU9025.
Thin air.
LP43699.
Third world.
MP25536.
This happy breed.
R593271.
This sporting life.
LF162.
This time for keeps.
R588088.
Thomas (Tom) Organization.
MP25720.
Thorne Films, Inc.
MP25856 - MP25874.
Three-cornered cage.
LP43800.
Three Musketeers.
LF190.
Three sisters.
LF184.
Threshold: the Blue Angels experience.
MP25920.
Thunder in the valley.
R583806.
Thunder Mountain.
R584161.
Tiara Tahiti.
LF168.
Tikki tikki tembo.
LP43684.
A Time for learning.
MP26010.
Time, Inc.
MP25747.
MP26035 - MP26047.
R584088 - R584098.
R587470.
Time of darkness.
LP43642.
Time to discover.
MP25750.
A Time to heal.
MP25919.
Tizzard, Edward G.
MP25750.
T-men in action.
MP26044.
Tobis Klangfilm.
R583554.
R584010.
Tobis Zeigt.
R584006.
R584008.
Tomorrow Entertainment, Inc.
LP43755.
Tomorrow’s Mexico.
R584090.
Tonight at 8:30.
LF199.
Tonometry.
MP25959.
Too long to wait.
LP43950.
Tools for cutting: stone axes to lasers.
MP25730.
Tooth or consequences.
R582964.
To Paris with love.
LF166
Top Choice.
MP25543.
Topol, John.
MU9037.
To ride a yeller horse.
LP43876.
To stalk the night.
LP43701.
Touchstone.
LP43928.
Touch your face.
MU9164.
The Town tamers.
LP43871.
Trailing danger.
R588501.
A Trail of bloodshed.
LP43869.
Trail of the lonesome Pink.
LP43766.
Trail to San Antone.
R588538.
Train, Middleton G. C.
MU9024.
Training the construction laborer.
MP25714.
Transition.
MP25565.
Transition-lift.
MP25564.
Transparent animals.
MP25660 - MP25663.
MP26057 - MP26058.
Trap that failed.
R587222.
The Treasure hunt (teamwork)
MP25624.
The Trespasser.
R582739.
Trevor, Elleston.
LF202.
Trial by knife.
LP43654.
Trial of Donald Duck.
R589193.
Triangles.
LP43894.
Tricks are not treats.
LP43819.
Triode vacuum tube voltmeters.
MP25525.
Tropical biology.
MP25811.
Tropical harmony.
R584821.
Trouble with women.
R587023.
Trustees of Trust under the Will of Mark J. Hellinger.
R585843.
Truth.
MP25749.
Try to die on time.
LP43823.
Tuesday movie of the week.
LP43716.
Tundra is a wide road.
MU9030.
Turkey’s 100 million.
MP26042.
Turn the key softly.
LF192.
Twentieth Century-Fox Film Corporation.
LF190.
LP43734.
LP43735.
LP43736.
LP43738.
LF43740.
LP43763.
LP43773.
LP43774.
LP43879.
LU3678.
R583794 - R583816.
R584187 - R584188.
R593725 - R593750.
Twenty-one cent bet and a positive slump eliminator.
MP25570.
Twenty thousand streets under the sky.
LF163.
Twilight on the Rio Grande.
R588539.
Two.
LP43909.
Two blondes and a redhead.
R591426.
Two Cities Films, Ltd.
LF158.
LF166.
LF178.
R590127.
R590128.
R593275.
U
UFA Filmkunst.
R583549.
R583552.
R585555.
R583556.
R584007.
R584559.
UFA Filmkunst Zeigt.
R584009.
U.F.O.’s.
LP43905.
Uncle Silas.
R593275.
Uncle Tom’s cabana.
R581040.
Unconquered.
R587026.
Under Colorado skies.
R590897.
Undercover agent.
MU9161.
Under the Tonto Rim.
R586033.
Unekis, Richard.
LP43734.
The Unfaithful.
R583122.
The Unfinished dance.
R583044.
Unicorn Productions, Inc.
MP25705.
Union Pacific Railroad.
MP25709.
Union Pacific Railroad. Department of Public Relations. Audio-
Visual Services.
MP25709.
Union Pacific Railroad. Safety Department.
MP25709.
United Artists.
LP43802.
United Artists Corporation.
LP43679.
LP43726.
LP43727.
LP43728.
LP43765.
LP43766.
LP43767.
LP43775.
LP43783.
LP43784.
LP43785.
LF43888.
LP43919.
LP43920.
LP43921.
LP43957.
United Artists Television, Inc.
R580300 - R580305.
R583122 - R583125.
R585005 - R585013.
R587017 - R587022.
R590795 - R590803.
R592812 - R592817.
United Brands Company.
R591937.
R592080 - R592088.
United States. Department of Justice. Law Enforcement Assistance
Administration.
MP25993 - MP25996.
United States. Department of the Interior. National Park Service.
MP25884.
United States Golf Association.
LP43794.
Units of conviction.
MP25981.
Universal.
LP43806 - LP43809.
Universal City Studios, Inc.
LP43697 - LP43705.
LP43707 - LP43708.
LP43710 - LP43725.
LP43756.
LP43804 - LP43805.
LP43932 - LP43938.
Universal City Studios, Inc. Emka Division.
R587023 - R587027.
R593022 - R593023.
Universal-International.
LF181.
Universal International Pictures, Inc.
R584818.
Universal Pictures.
LP43743.
LP43768.
LP43769.
LP43776.
LP43777.
R583192 - R583202.
R584814 - R584840.
R587707 - R587716.
R590970 - R590980.
R591746 - R591754.
Universal Pictures Company, Inc.
R583192.
R583194 - R583202.
R584814 - R584817.
R584819 - R584840.
R585843.
R587707 - R587716.
R590970 - R590980.
R591746 - R591754.
R593269 - R593273.
Universal Pictures, Ltd.
LP43770.
Universal Productions France, S.A.
LP43786.
Universal Studios.
LP43786.
University of Louisville (KY) National Crime Prevention Institute.
MP25812.
University of Michigan, Ann Arbor.
MP25965.
University of Michigan, Ann Arbor. Audio-Visual Education Center.
MP25965.
University of Michigan, Ann Arbor. Department of Electrical and
Computer Engineering.
MP25965.
University of Oregon, Portland. Medical School. Division of
Instructional Aids.
MP25775.
University of South Carolina, Columbia. Division of Educational
Services.
MP25562.
UniWorld Group, Inc.
MP25746.
Unsuspected.
R590796.
Unusual behavior patterns in fishes.
MP25874.
Up here looking down.
MP25614.
Up n’ atom.
R582972.
Uptown Saturday night.
LP43810.
Urban design issues in Texas.
MP25808.
Urchins of the coral reef.
MP25863.
Use of art therapy in a vocational milieu.
MP25825.
V
Vacant land.
MU9028.
Vacation days.
R584642.
Vacation magic.
R593732.
Vacations (two weeks a year)
R593749.
Valley of fear.
R585845.
Valse des toreadors.
LF160.
Vanished traces.
R584008.
Vantage Communications, Inc.
MP25675.
Variety girl.
R587027.
Velveteen rabbit.
LP43908.
Vera paints Ibiza in the sun.
LP26034
Verdon Productions, Ltd.
LP43810.
Verwehte Spuren.
R584008.
A Very missing person.
LP43724.
Very natural thing.
MP25546.
Viacom international, Inc.
R591488 - R591494.
Videocom, Inc.
MP25976 - MP25992.
View from the crib.
MP25819.
Vigilante.
R585734 - R585737.
R587216 - R587224.
Vigilantes return.
R584815.
Violence.
R591328.
Virginia Polytechnic Institute and State University, Blacksburg.
MP25820.
Virginia Polytechnic Institute and State University, Blacksburg.
College of Agriculture and Life Sciences.
MP25820.
Virginia Polytechnic Institute and State University, Blacksburg. Film
Unit.
MP25820.
Virtual storage concepts.
MP25582 - MP25585.
MP25737.
MP25952.
Vision Associates, Inc.
MP25558.
Visiting Virginia.
R589600.
The Visitor.
LP43979.
Visual Education Centre.
MP25893.
MP25894.
Visual Education Centre, a division of Public Media, Inc. SEE Public
Media, Inc. Visual Education Centre.
Visual Instruction Productions, a department of Victor Kayfetz
Productions, Inc. SEE Kayfetz (Victor) Productions, Inc. Visual
Instruction Productions.
Vitaphone Corporation.
R580303.
R580304.
R580305.
R583124.
R583125.
R585007 - R585013.
R587018.
R587021.
R587022.
R590798 - R590803.
R592815 - R592817.
Vivian’s problem.
LP43841.
Voice of the turtle.
R590797.
Volley oop.
R582974.
Voltmeter multiplier actions.
MP25518.
Volunteer.
MP25735.
Vortex, Inc.
LU3676.
VSAM concepts and access method services usage (DOS/VS)
MP25779.
VSAM macro coding and debugging.
MP25778.
W
Waber, Bernard.
LP43690.
Wade (Roger) Productions, Inc.
LP43890 - LP43903.
Walking home.
MU9153.
Walking safely to school.
MP26011.
Wall, Robert E.
MU9042.
Wallis (Hal) Productions, Inc.
R585723.
Walter Kerr on theater.
LP43883.
Walter’s holiday.
LP43835.
Waltz of the toreadors.
LF160.
W and W Films.
LP43794.
The Wankel engine—an analysis.
MP25688.
Wanted—more homes.
R584089.
Warner, Douglas.
LF207.
Warner Brothers.
LP43678.
LP43811.
LP43889.
Warner Brothers, Inc.
LP43741.
LP43744.
LP43745.
LP43757.
LP43764.
LP43778.
LP43787 - LP43791.
LP43810.
LP43812.
LP43813.
LP43913.
LP43939.
MP25503.
R583818.
R591573.
Warner Brothers Pictures Corporation.
R580302.
Warner Brothers Pictures, Inc.
R580301.
R583122.
R583123.
R585005.
R585006.
R587019.
R590797.
R592812.
R592813.
Warner Brothers Productions, Ltd.
LP43913.
Warner Communications.
LP43764.
Warner Communications Company.
LP43745.