Maximum Entropy Method Project

The goal of a project on the Maximum Entropy Method (MEM) and a project on multipole refinements of the IUCr Commission on Charge, Spin and Momentum Densities (CSMD) is to test the value of MEM for the study of CSMD. Objectives of the project include the following:

1. Survey the state of the discipline. Establish a system to gather information on recent results and computer programs.
2. Analyze the results. Is the same result obtained when a set of data is analyzed by different groups, using different programs and/or applying the programs in different ways? Establish criteria for weighting schemes, appropriate initial densities, and suitable experimental conditions for MEM analysis.
3. Identify the most efficient algorithms for MEM calculation and determine program limitations.
4. Analyze MEM densities in terms of models thus allowing a quantitative description of the underlying density. Characterize densities by topological  properties.
5. Determine how to estimate the accuracy of the MEM charge densities and avoid pitfalls and artifacts.

In order to guide the use of MEM, it is desirable to have a better understanding of its theoretical background. Although philosophical discussions of MEM are outside the scope of the project, it is hoped that the project will lead to better use of MEM in other applications as well.

Those interested in participating should contact a steering committee member: M. Sakata (Chair), a40366a@nucc.cc.nagoya-u.ac.jp; M. Takata (Sec.), a41024a@nucc.cc.nagoya-u.ac.jp; Members: F. K. Larsen (Co-chair), R. J. Papoular, W. Jauch; Observer: D. Feil; Advisors: G. Bricogne, D. M. Collins. The MEM calculation program MEED is available by anonymous ftp: (ftp://hod.nuap.nagoya-u.ac.jp/pub/ftp.html). There are no restrictions on methods of MEM data analysis or interpretation, and publication by participants will not be restricted.