|
|
Line 1: |
Line 1: |
| In [[mathematics]], the '''Bernoulli scheme''' or '''Bernoulli shift''' is a generalization of the [[Bernoulli process]] to more than two possible outcomes.<ref>P. Shields, ''The theory of Bernoulli shifts'' , Univ. Chicago Press (1973)</ref><ref>Michael S. Keane, "Ergodic theory and subshifts of finite type", (1991), appearing as Chapter 2 in ''Ergodic Theory, Symbolic Dynamics and Hyperbolic Spaces'', Tim Bedford, Michael Keane and Caroline Series, Eds. Oxford University Press, Oxford (1991). ISBN 0-19-853390-X</ref> Bernoulli schemes are important in the study of [[dynamical system]]s, as most such systems (such as [[Axiom A system]]s) exhibit a [[repellor]] that is the product of the [[Cantor set]] and a [[smooth manifold]], and the dynamics on the Cantor set are isomorphic to that of the Bernoulli shift.<ref>Pierre Gaspard, ''Chaos, scattering and statistical mechanics''(1998), Cambridge University press</ref> This is essentially the [[Markov partition]]. The term ''shift'' is in reference to the [[shift operator]], which may be used to study Bernoulli schemes. The [[Ornstein isomorphism theorem]]<ref>{{springer|author=D.S. Ornstein|title=Ornstein isomorphism theorem|id=Ornstein_isomorphism_theorem&oldid=17997}}</ref> shows that Bernoulli shifts are isomorphic when their [[Kolmogorov entropy|entropy]] is equal. Finite [[stationary stochastic process]]es are isomorphic to the Bernoulli shift; in this sense, Bernoulli shifts are [[universal property|universal]]. | | In [[mathematics]], a '''translation plane''' is a particular kind of [[projective plane]], as considered as a combinatorial object.<ref>Projective Planes [http://www.maths.qmul.ac.uk/~pjc/pps/pps2.pdf On projective planes]</ref> |
|
| |
|
| ==Definition==
| | In a projective plane, <math>\scriptstyle p</math> represents a point, and <math>\scriptstyle L</math> represents a line. A central [[collineation]] with center <math>\scriptstyle p</math> and axis <math>\scriptstyle L</math> is a collineation fixing every point on <math>\scriptstyle L</math> and every line through <math>\scriptstyle p</math>. It is called an "elation" if <math>\scriptstyle p</math> is on <math>\scriptstyle L</math>, otherwise it is called a "homology". The central collineations with centre <math>\scriptstyle p</math> and axis <math>\scriptstyle L</math> form a group.<ref>Geometry [http://www.math.uni-kiel.de/geometrie/klein/math/geometry/translation.html Translation Plane] Retrieved on June 13, 2007</ref> |
| A Bernoulli scheme is a [[discrete-time]] [[stochastic process]] where each [[statistical independence|independent]] [[random variable]] may take on one of ''N'' distinct possible values, with the outcome ''i'' occurring with probability <math>p_i</math>, with ''i'' = 1, ..., ''N'', and | |
|
| |
|
| :<math>\sum_{i=1}^N p_i = 1. </math>
| | A projective plane <math>\scriptstyle \Pi</math> is called a translation plane if there exists a line <math>\scriptstyle L</math> such that the group of elations with axis <math>\scriptstyle L</math> is transitive on the affine plane Π<sub>l</sub> (the [[Affine geometry|affine]] derivative of Π). |
|
| |
|
| The [[sample space]] is usually denoted as
| | == Relationship to spreads == |
| | Translation planes are related to spreads in finite projective spaces by the André/Bruck-Bose construction.<ref>{{cite web|url=http://www-ma4.upc.es/~simeon/bblpsympspread.pdf|title=Symplectice Spreads|last=Ball|first=Simeon|author2=John Bamberg |author3=Michel Lavrauw |author4=Tim Penttila |date=2003-09-15|publisher=[[Polytechnic University of Catalonia]]|accessdate=2008-10-08}}</ref> A spread of <math>\scriptstyle PG(3, q) </math> is a set of ''q''<sup>2</sup> + 1 lines, with no two intersecting. Equivalently, it is a partition of the points of <math>\scriptstyle PG(3, q) </math> into lines. |
|
| |
|
| :<math>X=\{1,\ldots,N \}^\mathbb{Z}</math>
| | Given a spread <math>\scriptstyle S</math> of <math>\scriptstyle PG(3, q) </math>, the André/Bruck-Bose construction<sup>1</sup> produces a translation plane <math>\scriptstyle \pi(S)</math> of order ''q''<sup>2</sup> as follows: Embed <math>\scriptstyle PG(3, q) </math> as a hyperplane of <math>\scriptstyle PG(4, q) </math>. Define an incidence structure <math>\scriptstyle A(S)</math> with "points," the points of <math>\scriptstyle PG(4, q) </math> not on <math>\scriptstyle PG(3, q) </math> and "lines" the planes of <math>\scriptstyle PG(4, q) </math> meeting <math>\scriptstyle PG(3, q) </math> in a line of <math>\scriptstyle S</math>. Then <math>\scriptstyle A(S)</math> is a translation affine plane of order ''q''<sup>2</sup>. Let <math>\scriptstyle \pi(S)</math> be the projective completion of <math>\scriptstyle A(S)</math>.<ref>{{cite book |
| | | | last =André | first =Johannes | authorlink = | title = Über nicht-Dessarguessche Ebenen mit transitiver Translationsgruppe | publisher = | year =1954 | location = | pages =156–186 | url = | doi = | id = }}</ref><ref>{{cite book |
| as a short-hand for
| | | last =Bruck | first = R. H. | authorlink = Richard Bruck|author2=R. C. Bose | title = The Construction of Translation Planes from Projective Spaces | publisher = | year =1964 | location = | pages = 85–102 | url = | doi = | id = }}</ref> |
| | |
| :<math>X=\{ x=(\ldots,x_{-1},x_0,x_1,\ldots) :
| |
| x_k \in \{1,\ldots,N\} \; \forall k \in \mathbb{Z} \}.</math>
| |
| | |
| The associated [[measure (mathematics)|measure]] is called the '''Bernoulli measure'''<ref>Achim Klenke, ''Probability Theory'' (2006) Springer-Verlag, ISBN 978-1-848000-047-6 doi:10.1007/978-1-848000-048-3</ref>
| |
| | |
| :<math>\mu = \{p_1,\ldots,p_N\}^\mathbb{Z}</math>
| |
| | |
| The [[sigma-algebra|σ-algebra]] <math>\mathcal{A}</math> on ''X'' is the product sigma algebra; that is, it is the (countable) [[direct product]] of the σ-algebras of the finite set {1, ..., ''N''}. Thus, the triplet
| |
| | |
| :<math>(X,\mathcal{A},\mu)</math>
| |
| | |
| is a [[measure space]]. The elements of <math>\mathcal{A}</math> are commonly called [[cylinder set]]s. Given a cylinder set <math>[x_0, x_1,\ldots,x_n]</math>, its measure is
| |
| | |
| :<math>\mu\left([x_0, x_1,\ldots,x_n]\right)=
| |
| \prod_{i=0}^n p_{x_i}</math>
| |
| The equivalent expression, using the notation of probability theory, is
| |
| :<math>\mu\left([x_0, x_1,\ldots,x_n]\right)=
| |
| \mathrm{Pr}(X_0=x_0, X_1=x_1, \ldots, X_n=x_n)</math>
| |
| for the random variables <math>\{X_k\}</math>
| |
| | |
| The Bernoulli scheme, as any stochastic process, may be viewed as a [[dynamical system]] by endowing it with the [[shift operator]] ''T'' where
| |
| | |
| :<math>Tx_k = x_{k+1}.</math>
| |
| | |
| Since the outcomes are independent, the shift preserves the measure, and thus ''T'' is a [[measure-preserving transformation]]. The quadruplet
| |
| | |
| :<math>(X,\mathcal{A},\mu, T)</math>
| |
| | |
| is a [[measure-preserving dynamical system]], and is called a '''Bernoulli scheme''' or a '''Bernoulli shift'''. It is often denoted by
| |
| | |
| :<math>BS(p)=BS(p_1,\ldots,p_N).</math>
| |
| | |
| The ''N'' = 2 Bernoulli scheme is called a [[Bernoulli process]]. The Bernoulli shift can be understood as a special case of the [[Markov shift]], where all entries in the [[adjacency matrix]] are one, the corresponding graph thus being a [[clique (graph theory)|clique]].
| |
| | |
| ==Generalizations==
| |
| Most of the properties of the Bernoulli scheme follow from the countable [[direct product]], rather than from the finite base space. Thus, one may take the base space to be any [[standard probability space]] <math>(Y,\mathcal{B},\nu)</math>, and define the Bernoulli scheme as
| |
| :<math>(X, \mathcal{A}, \mu)=(Y,\mathcal{B},\nu)^\mathbb{Z}</math>
| |
| This works because the countable direct product of a standard probability space is again a standard probability space.
| |
| | |
| As a further generalization, one may replace the in integers <math>\mathbb{Z}</math> by a [[countable]] [[discrete group]] <math>G</math>, so that
| |
| :<math>(X, \mathcal{A}, \mu)=(Y,\mathcal{B},\nu)^G</math>
| |
| For this last case, the shift operator is replaced by the [[group action]]
| |
| :<math>gx(f)=x(g^{-1}f)</math>
| |
| for group elements <math>f,g\in G</math> and <math>x\in Y^G</math> understood as a function <math>x:G\to Y</math> (any direct product <math>Y^G</math> can be understood to be the set of functions <math>[G\to Y]</math>, as this is the [[exponential object]]). The measure <math>\mu</math> is taken as the [[Haar measure]], which is invariant under the group action:
| |
| :<math>\mu(gx)=\mu(x). \, </math>
| |
| These generalizations are also commonly called Bernoulli schemes, as they still share most properties with the finite case.
| |
| | |
| ==Properties==
| |
| [[Ya. Sinai]] demonstrated that the [[Kolmogorov entropy]] of a Bernoulli scheme is given by<ref>Ya.G. Sinai, (1959) "On the Notion of Entropy of a Dynamical System", ''Doklady of Russian Academy of Sciences'' '''124''', pp. 768–771.</ref><ref>Ya. G. Sinai, (2007) "[http://web.math.princeton.edu/facultypapers/Sinai/MetricEntropy2.pdf Metric Entropy of Dynamical System]"</ref>
| |
| | |
| :<math>H = -\sum_{i=1}^N p_i \log p_i .</math>
| |
| | |
| This may be seen as resulting from the general definition of the entropy of a [[Cartesian product]] of probability spaces, which follows from the [[asymptotic equipartition property]]. For the case of a general base space <math>(Y, \mathcal{B}, \nu)</math> (''i.e.'' a base space which is not countable), one typically considers the [[relative entropy]]. So, for example, if one has a countable [[partition of a set|partition]] <math>Y'\subset Y</math> of the base ''Y'', such that <math>\nu(Y')=1</math>, one may define the entropy as
| |
| | |
| :<math>H_{Y'} = -\sum_{y'\in Y'} \nu(y') \log \nu(y') .</math>
| |
| | |
| In general, this entropy will depend on the partition; however, for many [[dynamical system]]s, it is the case that the [[symbolic dynamics]] is independent of the partition (or rather, there are isomorphisms connecting the symbolic dynamics of different partitions, leaving the measure invariant), and so such systems can have a well-defined entropy independent of the partition.
| |
| | |
| The [[Ornstein isomorphism theorem]] states that two Bernoulli schemes with the same entropy are [[isomorphism of dynamical systems|isomorphic]].<ref>Donald Ornstein, "Bernoulli shifts with the same entropy are isomorphic", ''Advances in Math.'' '''4''' (1970), pp.337–352</ref> The result is sharp<ref>Christopher Hoffman, "[http://www.ams.org/journals/tran/1999-351-10/S0002-9947-99-02446-0/ A K counterexample machine]", ''Trans. Amer. Math. Soc.'' '''351''' (1999), pp 4263–4280 </ref>, in that very similar, non-scheme systems, such as [[Kolmogorov automorphism]]s, do not have this property.
| |
| | |
| The Ornstein isomorphism theorem is in fact considerably deeper: it provides a simple criterion by which many different [[measure-preserving dynamical system]]s can be judged to be isomorphic to Bernoulli schemes. The result was surprising, as many systems previously believed to be unrelated proved to be isomorphic. These include all finite{{clarify|date=November 2010}} [[stationary stochastic process]]es, [[subshifts of finite type]], finite [[Markov chain]]s, [[Anosov flow]]s, and [[Sinai's billiards]]: these are all isomorphic to Bernoulli schemes.
| |
| | |
| For the generalized case, the Ornstein isomorpism theorem still holds if the group ''G'' is a countably infinite [[amenable group]].
| |
| <ref>D. Ornstein and B. Weiss. "Entropy and isomorphism theorems for actions of amenable groups." ''J. Analyse Math.'' '''48''' (1987), pp. 1–141. </ref><ref>Lewis Bowen (2011), "[http://arxiv.org/abs/1103.4424 Every countably infinite group is almost Ornstein]", ArXiv abs/1103.4424</ref>
| |
| | |
| ==Bernoulli automorphism== | |
| An invertible, [[measure-preserving transformation]] of a [[standard probability space]] (Lebesgue space) is called a '''Bernoulli automorphism''' if it [[isomorphism of dynamical systems|isomorphic]] to a Bernoulli shift.<ref>Peter Walters (1982) ''An Introduction to Ergodic Theory'', Springer-Verlag, ISBN 0-387-90599-5</ref>
| |
| | |
| ==See also==
| |
| * [[Shift of finite type]]
| |
| * [[Markov chain]]
| |
| * [[Hidden Bernoulli model]]
| |
|
| |
|
| ==References== | | ==References== |
| <references/>
| | {{Reflist}} |
|
| |
|
| {{DEFAULTSORT:Bernoulli Scheme}}
| | ==Further reading== |
| [[Category:Markov models]] | | * Mauro Biliotti, Vikram Jha, Norman L. Johnson (2001) ''Foundations of Translation Planes'', [[Marcel Dekker]] ISBN 0-8247-0609-9 . |
| [[Category:Ergodic theory]]
| |
| [[Category:Stochastic processes]]
| |
| [[Category:Symbolic dynamics]]
| |
|
| |
|
| | ==External links== |
| | *[http://www.library.tuiasi.ro/ipm/vol13no34/pure.html Foundations_of_Translation_Planes] |
| | *[http://www-math.ucdenver.edu/~wcherowi/courses/m6221/pglc3a.html Lecture Notes on Projective Geometry] |
| | *[http://mellinger.umwblogs.org/publications/ Publications of Keith Mellinger] |
|
| |
|
| [[fr:Décalage de Bernoulli (langage formel)]]
| | {{DEFAULTSORT:Translation Plane}} |
| [[ru:Символическая динамика]] | | [[Category:Projective geometry]] |
In mathematics, a translation plane is a particular kind of projective plane, as considered as a combinatorial object.[1]
In a projective plane, represents a point, and represents a line. A central collineation with center and axis is a collineation fixing every point on and every line through . It is called an "elation" if is on , otherwise it is called a "homology". The central collineations with centre and axis form a group.[2]
A projective plane is called a translation plane if there exists a line such that the group of elations with axis is transitive on the affine plane Πl (the affine derivative of Π).
Relationship to spreads
Translation planes are related to spreads in finite projective spaces by the André/Bruck-Bose construction.[3] A spread of is a set of q2 + 1 lines, with no two intersecting. Equivalently, it is a partition of the points of into lines.
Given a spread of , the André/Bruck-Bose construction1 produces a translation plane of order q2 as follows: Embed as a hyperplane of . Define an incidence structure with "points," the points of not on and "lines" the planes of meeting in a line of . Then is a translation affine plane of order q2. Let be the projective completion of .[4][5]
References
43 year old Petroleum Engineer Harry from Deep River, usually spends time with hobbies and interests like renting movies, property developers in singapore new condominium and vehicle racing. Constantly enjoys going to destinations like Camino Real de Tierra Adentro.
Further reading
- Mauro Biliotti, Vikram Jha, Norman L. Johnson (2001) Foundations of Translation Planes, Marcel Dekker ISBN 0-8247-0609-9 .
External links