-
On a Proof of the Convergence Speed of a Second-order Recurrence Formula in the Arimoto-Blahut Algorithm
Authors:
Kenji Nakagawa,
Yoshinori Takei,
Shin-ichiro Hara
Abstract:
In [8] (Nakagawa, et.al., IEEE Trans. IT, 2021), we investigated the convergence speed of the Arimoto-Blahut algorithm. In [8], the convergence of the order $O(1/N)$ was analyzed by focusing on the second-order nonlinear recurrence formula consisting of the first- and second-order terms of the Taylor expansion of the defining function of the Arimoto-Blahut algorithm. However, in [8], an infinite n…
▽ More
In [8] (Nakagawa, et.al., IEEE Trans. IT, 2021), we investigated the convergence speed of the Arimoto-Blahut algorithm. In [8], the convergence of the order $O(1/N)$ was analyzed by focusing on the second-order nonlinear recurrence formula consisting of the first- and second-order terms of the Taylor expansion of the defining function of the Arimoto-Blahut algorithm. However, in [8], an infinite number of inequalities were assumed as a "conjecture," and proofs were given based on the conjecture. In this paper, we report a proof of the convergence of the order $O(1/N)$ for a class of channel matrices without assuming the conjecture. The correctness of the proof will be confirmed by several numerical examples.
△ Less
Submitted 11 September, 2022;
originally announced September 2022.
-
Continuity of isomorphisms applied to rigidity problems of entropy spectra
Authors:
Katsukuni Nakagawa
Abstract:
For a fixed topological Markov shift, we consider measure-preserving dynamical systems of Gibbs measures for 2-locally constant functions on the shift. We also consider isomorphisms between two such systems. We study the set of all 2-locally constant functions $f$ on the shift such that all those isomorphisms defined on the system associated with $f$ are induced from automorphisms of the shift. We…
▽ More
For a fixed topological Markov shift, we consider measure-preserving dynamical systems of Gibbs measures for 2-locally constant functions on the shift. We also consider isomorphisms between two such systems. We study the set of all 2-locally constant functions $f$ on the shift such that all those isomorphisms defined on the system associated with $f$ are induced from automorphisms of the shift. We prove that this set contains a full-measure open set of the space of all 2-locally constant functions on the shift. We apply this result to rigidity problems of entropy spectra and show that the strong non-rigidity occurs if and only if so does the weak non-rigidity.
△ Less
Submitted 28 October, 2021; v1 submitted 11 October, 2021;
originally announced October 2021.
-
Weak rigidity of entropy spectra
Authors:
Katsukuni Nakagawa
Abstract:
In this paper, we consider entropy spectra on topological Markov shifts. We prove that if two measure-preserving dynamical systems of Gibbs measures with Hölder continuous potentials are isomorphic, then their entropy spectra are the same. This result raises a new rigidity problem. We call this problem the weak rigidity problem, contrasting it with the strong rigidity problem proposed by Barreira…
▽ More
In this paper, we consider entropy spectra on topological Markov shifts. We prove that if two measure-preserving dynamical systems of Gibbs measures with Hölder continuous potentials are isomorphic, then their entropy spectra are the same. This result raises a new rigidity problem. We call this problem the weak rigidity problem, contrasting it with the strong rigidity problem proposed by Barreira and Saraiva. We give a complete answer to the weak rigidity problem for Markov measures on a topological Markov shift with an aperiodic transition matrix of size 2.
△ Less
Submitted 25 September, 2020; v1 submitted 14 September, 2020;
originally announced September 2020.
-
Decomposition of Normal Operators and Its Application to Spectral Theorem
Authors:
Katsukuni Nakagawa
Abstract:
A decomposition theorem for self-adjoint operators proved by Riesz and Lorch is extended to normal operators. This extension gives a new proof of the spectral theorem for unbounded normal operators.
A decomposition theorem for self-adjoint operators proved by Riesz and Lorch is extended to normal operators. This extension gives a new proof of the spectral theorem for unbounded normal operators.
△ Less
Submitted 28 June, 2020; v1 submitted 8 June, 2020;
originally announced June 2020.
-
Compactness of Transfer Operators and Spectral Representation of Ruelle Zeta Functions for Super-continuous Functions
Authors:
Katsukuni Nakagawa
Abstract:
Transfer operators and Ruelle zeta functions for super-continuous functions on one-sided topological Markov shifts are considered. For every super-continuous function, we construct a Banach space on which the associated transfer operator is compact. Using this Banach space, we establish the trace formula and spectral representation of Ruelle zeta functions for a certain class of super-continuous f…
▽ More
Transfer operators and Ruelle zeta functions for super-continuous functions on one-sided topological Markov shifts are considered. For every super-continuous function, we construct a Banach space on which the associated transfer operator is compact. Using this Banach space, we establish the trace formula and spectral representation of Ruelle zeta functions for a certain class of super-continuous functions. Our results include, as a special case, the classical trace formula and spectral representation for the class of locally constant functions.
△ Less
Submitted 16 June, 2020; v1 submitted 2 June, 2020;
originally announced June 2020.
-
TPLVM: Portfolio Construction by Student's $t$-process Latent Variable Model
Authors:
Yusuke Uchiyama,
Kei Nakagawa
Abstract:
Optimal asset allocation is a key topic in modern finance theory. To realize the optimal asset allocation on investor's risk aversion, various portfolio construction methods have been proposed. Recently, the applications of machine learning are rapidly growing in the area of finance. In this article, we propose the Student's $t$-process latent variable model (TPLVM) to describe non-Gaussian fluctu…
▽ More
Optimal asset allocation is a key topic in modern finance theory. To realize the optimal asset allocation on investor's risk aversion, various portfolio construction methods have been proposed. Recently, the applications of machine learning are rapidly growing in the area of finance. In this article, we propose the Student's $t$-process latent variable model (TPLVM) to describe non-Gaussian fluctuations of financial timeseries by lower dimensional latent variables. Subsequently, we apply the TPLVM to minimum-variance portfolio as an alternative of existing nonlinear factor models. To test the performance of the proposed portfolio, we construct minimum-variance portfolios of global stock market indices based on the TPLVM or Gaussian process latent variable model. By comparing these portfolios, we confirm the proposed portfolio outperforms that of the existing Gaussian process latent variable model.
△ Less
Submitted 28 January, 2020;
originally announced February 2020.
-
Tail Probability and Singularity of Laplace-Stieltjes Transform of a Heavy Tailed Random Variable
Authors:
Kenji Nakagawa
Abstract:
In this paper, we will give a sufficient condition for a non-negative random variable $X$ to be heavy tailed by investigating the Laplace-Stieltjes transform of the probability distribution function. We focus on the relation between the singularity at the real point of the axis of convergence and the asymptotic decay of the tail probability. Our theorem is a kind of Tauberian theorems.
In this paper, we will give a sufficient condition for a non-negative random variable $X$ to be heavy tailed by investigating the Laplace-Stieltjes transform of the probability distribution function. We focus on the relation between the singularity at the real point of the axis of convergence and the asymptotic decay of the tail probability. Our theorem is a kind of Tauberian theorems.
△ Less
Submitted 1 September, 2009; v1 submitted 1 September, 2009;
originally announced September 2009.
-
On non-eliminability of the cut rule and the roles of associativity and distributivity in non-commutative substructural logics
Authors:
Takeshi Ueno,
Koji Nakaogawa,
Osamu Watari
Abstract:
We introduce a sequent calculus FL' for non-commutative substructural logic. It has at most one formula on the right side of sequent, and excludes three structural inference rules, i.e. contraction, weakening and exchange. (FL' is based on our investigations of the Gentzen-style natural deduction for non-commutative substructural logics.) FL' has the same proof strength as the standard sequent c…
▽ More
We introduce a sequent calculus FL' for non-commutative substructural logic. It has at most one formula on the right side of sequent, and excludes three structural inference rules, i.e. contraction, weakening and exchange. (FL' is based on our investigations of the Gentzen-style natural deduction for non-commutative substructural logics.) FL' has the same proof strength as the standard sequent calculus FL (Full Lambek), which is the basic sequent calculus for all other substructural logics. For the standard FL, we use Ono's formulation. Although FL' and the standard FL are equivalent, there is a subtle difference in the left rule of implication. In the standard FL, two parameters $Γ_1$ and $Γ_2$(resp.), each of which is just an finite sequence of arbitrary formulas, appear on the left and right side (resp.) of a formula which is placed on the left side of the sequent on the upper left side of the left rule $\imply$ (which corresponds to $\imply'$ in FL'). On the other hand, there is no such parameter on the left side of the sequent on the upper left side in the left rule for $\imply'$ of FL'. In FL', $Γ_1$ is always empty, and only $Γ_2$ is allowed to occur in the left rule for $\imply'$. (Similar differences occur in multiplicative and additive conjunctions, and in additive disjunction.) This difference between FL' and FL matters deeply, for we are led to a construction of proof-figures in FL', which show how the associativity of multiplicative conjunction and the distributivity of multiplicative conjunction over additive disjunction are involved in the eliminations of the cut rule in those proof-figures. We specify how associativity and distributivity are related to the non-eliminability of an application of the cut rule in those proof-figures of FL'.
△ Less
Submitted 2 February, 2009;
originally announced February 2009.
-
Application of Tauberian Theorem to the Exponential Decay of the Tail Probability of a Random Variable
Authors:
Kenji Nakagawa
Abstract:
We give a sufficient condition for the exponential decay of the tail probability of a non-negative random variable. We consider the Laplace-Stieltjes transform of the probability distribution function of the random variable. We present a theorem, according to which if the abscissa of convergence of the LS transform is negative finite and the real point on the axis of convergence is a pole of the…
▽ More
We give a sufficient condition for the exponential decay of the tail probability of a non-negative random variable. We consider the Laplace-Stieltjes transform of the probability distribution function of the random variable. We present a theorem, according to which if the abscissa of convergence of the LS transform is negative finite and the real point on the axis of convergence is a pole of the LS transform, then the tail probability decays exponentially. For the proof of the theorem, we extend and apply so-called a finite form of Ikehara's complex Tauberian theorem by Graham-Vaaler.
△ Less
Submitted 25 April, 2007; v1 submitted 25 February, 2006;
originally announced February 2006.
