. A C ] 1 6 Ju n 20 03 RIGID RESOLUTIONS AND BIG BETTI NUMBERS
نویسنده
چکیده
Let K be a field of characteristic 0 and S = K[x1, . . . , xn] the polynomial ring over K with graded maximal ideal m = (x1, . . . , xn). Denote by βi(M) = Tor S i (K,M) the ith Betti number of a finitely generated graded module M and by Gin(I) the generic initial ideal of a graded ideal I with respect to the reverse lexicographical order. In this paper we answer (positively) a question raised by the first author in [7]. We prove that if a graded ideal I ⊂ S has βi(I) = βi(Gin(I)) for some i, then βk(I) = βk(Gin(I)) for all k ≥ i, see Corollary 2.4. For i = 0, this theorem was first proved by Aramova, Herzog and Hibi [1]. More generally, we show that the same statement holds if Gin(I) is replaced by either any generic initial ideal of I or by the lex-segment ideal associated with I. Given a finitely generated graded S-module M , a generic sequence of linear forms y1, . . . , yn and an integer p, 1 ≤ p ≤ n, we define the generic annihilator number αp(M) of M to be dimK((y1, . . . , yp−1)M :M yp/(y1, . . . , yp−1)M and the generic Koszul homology Hi(y1, . . . , yp;M) to be the ith-homology of the Koszul complex over M with respect to y1, . . . , yp. In the first section of this paper we show that there is an upper bound for the Betti numbers of M in terms of the generic annihilator numbers of M . We show in Theorem 1.5, that among other equivalent conditions, this upper bound is achieved for all i if and only if mHi(y1, . . . , yp;M) = 0 for all i > 0 and p = 1, . . . , n. The above mentioned Corollary 2.4 is a consequence of the more general Theorem 2.3, proved in Section 2, which says that if the ith Betti number of M achieves the upper bound given by the generic annihilator numbers, then the upper bound is also achieved for jth Betti numbers with j > i. For the proof of this theorem the following interesting annihilation property of Koszul homology is required: suppose that for a generic sequence y = y1, . . . , yn of linear forms and some i one has mHi(y1, . . . , yp;M) = 0 for all p then mHk(y1, . . . , yp;M) = 0 for all k ≥ i and for all p. In the last section of the paper we show that if two componentwise linear ideals I ⊂ J ⊂ S have the same Hilbert polynomial then βi(J) ≤ βi(I) for all i, see Theorem 3.2. This theorem was inspired by a question of Eisenbud and Huneke: suppose char(K) = 0 and I is a graded m-primary ideal in S with I ⊂ m for some d. Is it then true that the number of generators of Gin(I) is greater than or equal to the number of generators of m? As an application of Theorem 3.2 we show in Corollary 3.4 that this is indeed the case. Moreover, we show that Gin(I) and m
منابع مشابه
ar X iv : a st ro - p h / 94 11 03 8 v 1 9 N ov 1 99 4 THE PARTICLE - AND ASTRO - PHYSICS OF DARK MATTER ⋆
We review some recent determinations of the amount of dark matter on galactic, cluster, and large scales, noting some puzzles and their possible resolutions. We discuss the interpretation of big bang nucleosynthesis for dark matter, and then review the motivation for and basic physics of several dark matter candidates, including Machos, Wimps, axions, and neutrinos. Finally, we discuss how the ...
متن کاملEfficient community identification and maintenance at multiple resolutions on distributed datastores
Article history: Received 16 September 2014 Received in revised form 6 May 2015 Accepted 2 June 2015 Available online 16 June 2015 The topic of network community identification at multiple resolutions is of great interest in practice to learn high cohesive subnetworks about different subjects in a network. For instance, one might examine the interconnections among web pages, blogs and social co...
متن کاملStructure Preserving Large Imagery Reconstruction
With the explosive growth of web-based cameras and mobile devices, billions of photographs are uploaded to the internet. We can trivially collect a huge number of photo streams for various goals, such as image clustering, 3D scene reconstruction, and other big data applications. However, such tasks are not easy due to the fact the retrieved photos can have large variations in their view perspec...
متن کاملDesign of Algorithms for Big Data Analytics
Very large volumes of data are being collected in every domain of human endeavour. The term Big Data has become pervasive recently and denotes the problems related to the storage, management, and analysis of these large amounts of data. Business, sciences, and engineering are all becoming dependent on their ability to harness useful knowledge from very large datasets. For example, the research ...
متن کاملProc. of the 2017 conference on Big Data from Space
A characteristic task in remote sensing Earth observation involves the registration of changes which may signal environmentally significant events. The Sentinel-1 synthetic aperture radar (SAR) and the Sentinel-2 optical/visible-infrared space-borne platforms, with spatial resolutions of the order of 10-20 meters and revisit times of the order of days, provide an attractive source of data for c...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2003