Showing 1–5 of 5 results for author: Grendar, M

Gibbs conditioning extended, Boltzmann conditioning introduced

Authors: Marian Grendar

Abstract: Conditional Equi-concentration of Types on I-projections (ICET) and Extended Gibbs Conditioning Principle (EGCP) provide an extension of Conditioned Weak Law of Large Numbers and of Gibbs Conditioning Principle to the case of non-unique Relative Entropy Maximizing (REM) distribution (aka I-projection). ICET and EGCP give a probabilistic justification to REM under rather general conditions. mu-pr… ▽ More Conditional Equi-concentration of Types on I-projections (ICET) and Extended Gibbs Conditioning Principle (EGCP) provide an extension of Conditioned Weak Law of Large Numbers and of Gibbs Conditioning Principle to the case of non-unique Relative Entropy Maximizing (REM) distribution (aka I-projection). ICET and EGCP give a probabilistic justification to REM under rather general conditions. mu-projection variants of the results are introduced. They provide a probabilistic justification to Maximum Probability (MaxProb) method. 'REM/MaxEnt or MaxProb?' question is discussed, briefly. Jeffreys Conditioning Principle is mentioned. △ Less

Submitted 17 September, 2004; v1 submitted 7 July, 2004; originally announced July 2004.

Comments: Three major changes: 1) Definition of proper I-projection has been changed. 2) An argument preceding Eq. (7) at the proof of ICET is now correctly stated. 3) Abstract was rewritten. To appear at Proceedings of MaxEnt 2004 workshop

MSC Class: 60F05; 60F10

Randomness as an Equilibrium. Potential and Probability Density

Authors: M. Grendar, Jr., M. Grendar

Abstract: Randomness is viewed through an analogy between a physical quantity, density of gas, and a mathematical construct -- probability density. Boltzmann's deduction of equilibrium distribution of ideal gas placed in an external potential field than provides a way of viewing probability density from a perspective of forces/potentials, hidden behind it. Randomness is viewed through an analogy between a physical quantity, density of gas, and a mathematical construct -- probability density. Boltzmann's deduction of equilibrium distribution of ideal gas placed in an external potential field than provides a way of viewing probability density from a perspective of forces/potentials, hidden behind it. △ Less

Submitted 2 December, 2002; originally announced December 2002.

Comments: 4 pages, MaxEnt 2001 workshop

MSC Class: 60A99

Journal ref: In: Bayesian inference and Maximum Entropy methods in Science and Engineering, R. L. Fry (ed.), AIP (Melville), 405-410, 2002

Why Maximum Entropy? A Non-axiomatic Approach

Authors: M. Grendar, Jr., M. Grendar

Abstract: Ill-posed inverse problems of the form y = X p where y is J-dimensional vector of a data, p is m-dimensional probability vector which cannot be measured directly and matrix X of observable variables is a known J,m matrix, J < m, are frequently solved by Shannon's entropy maximization (MaxEnt). Several axiomatizations were proposed to justify the MaxEnt method (also) in this context. The main aim… ▽ More Ill-posed inverse problems of the form y = X p where y is J-dimensional vector of a data, p is m-dimensional probability vector which cannot be measured directly and matrix X of observable variables is a known J,m matrix, J < m, are frequently solved by Shannon's entropy maximization (MaxEnt). Several axiomatizations were proposed to justify the MaxEnt method (also) in this context. The main aim of the presented work is two-fold: 1) to view the concept of complementarity of MaxEnt and Maximum Likelihood (ML) tasks from a geometric perspective, and consequently 2) to provide an intuitive and non-axiomatic answer to the 'Why MaxEnt?' question. △ Less

Submitted 2 December, 2002; originally announced December 2002.

Comments: 4 pages, MaxEnt 2001

MSC Class: 62B10; 94A17

Journal ref: In: Bayesian inference and Maximum Entropy methods in Science and Engineering, R. L. Fry (ed.), AIP (Melville), 375-379, 2002

Minimax Entropy and Maximum Likelihood. Complementarity of tasks, identity of solutions

Authors: Marian Grendar, Jr, Marian Grendar

Abstract: Concept of exponential family is generalized by simple and general exponential form. Simple and general potential are introduced. Maximum Entropy and Maximum Likelihood tasks are defined. ML task on the simple exponential form and ME task on the simple potentials are proved to be complementary in set-up and identical in solutions. ML task on the general exponential form and ME task on the genera… ▽ More Concept of exponential family is generalized by simple and general exponential form. Simple and general potential are introduced. Maximum Entropy and Maximum Likelihood tasks are defined. ML task on the simple exponential form and ME task on the simple potentials are proved to be complementary in set-up and identical in solutions. ML task on the general exponential form and ME task on the general potentials are weakly complementary, leading to the same necessary conditions. A hypothesis about complementarity of ML and MiniMax Entropy tasks and identity of their solutions, brought up by a special case analytical as well as several numerical investigations, is suggested in this case. MiniMax Ent can be viewed as a generalization of MaxEnt for parametric linear inverse problems, and its complementarity with ML as yet another argument in favor of Shannon's entropy criterion. △ Less

Submitted 13 September, 2000; originally announced September 2000.

Comments: Presented at 'MaxEnt 2000', CNRS, Gif sur Yvette, France, July 8-13 2000

MSC Class: 62A10; 62A99

Journal ref: In: Bayesian inference and Maximum Entropy methods in Science and Engineering, A. Mohammad-Djafari (ed.), AIP (Melville), pp. 49-61, 2001

What is the question that MaxEnt answers? A probabilistic interpretation

Authors: Marian Grendar Jr, Marian Grendar

Abstract: The Boltzmann-Wallis-Jaynes' multiplicity argument is taken up and elaborated. MaxEnt is proved and demonstrated to be just an asymptotic case of looking for such a vector of absolute frequencies in a feasible set, which has maximal probability of being generated by a uniform prior generator/pmf. The Boltzmann-Wallis-Jaynes' multiplicity argument is taken up and elaborated. MaxEnt is proved and demonstrated to be just an asymptotic case of looking for such a vector of absolute frequencies in a feasible set, which has maximal probability of being generated by a uniform prior generator/pmf. △ Less

Submitted 13 September, 2000; originally announced September 2000.

Comments: Presented at 'MaxEnt 2000', CNRS, Gif-sur-Yvette, France, July 8-13, 2000

MSC Class: 60C05 (primary); 60F99 (secondary); 62A15 (secondary)

Journal ref: In: Bayesian inference and Maximum Entropy methods in Science and Engineering, A. Mohammad-Djafari (ed.), AIP (Melville), pp. 83-94, 2001