Simulation Books

If you want explore the world of VR, this book is a good starting point!

I really enjoyed playing Compete. It gave my group and me to compete with other groups. We had to analyze financial statements and observe graphs. We'd enter prices and the number of units to be sold in each region etc. We had to be careful about everything we did. One mistake could ruin everything. It was very exciting and interesting. I'd definitely recommend it.

As a field of applied mathematics, computational biology has exploded in the last decade, and shows every sign of increasing in the next. This book overviews a few of the topics in the computational modeling of cells. I only read chapters 12 and 13 on molecular motors, and so my review will be confined to these.

Nanotechnology could be described as an up-and-coming field, but in the natural world one can find examples of this technology that surpass greatly what has been accomplished by human engineers. The authors begin their articles with a few examples of natural molecular machines, including the "rotary motors" DNA helicase and bacteriophage, and the "linear motor" kinesin, the latter they refer to as a "walking enzyme". Important in the modeling of all these is the theory of stochastic processes in the guise of Brownian motion, which the authors hold is the key to understanding the mechanics of proteins. In chapter 12 they give a detailed overview of the mathematical modeling of protein dynamics, followed in chapter 13 by an illustration of the mathematical formalism in the bacterial flagellar motor, a polymerization ratchet, and a motor governing ATP synthase.

To the authors a molecular motor is an entity that converts chemical energy into mechanical force. The production of mechanical force though may involve intermediate steps of energy transduction, all these involving the release of free energy during binding events. But due to their size, molecular motors are subjected to thermal fluctuations, and thus to model their motion accurately requires the theory of stochastic processes. Thus the authors begin a study of stochastic processes, restricting their attention to ones that satisfy the Markov property. Starting with a discrete model of protein motion as a simple random walk, the authors show that the variance of the motion grows linearly with time, which is a sign of diffusive motion. The partial differential equation satisfied by the probability distribution function, in the continuous limit where the space and time scales are large enough, is left to the reader to derive as an exercise.

The authors then consider polymer growth as another example of a stochastic process, a kind of hybrid one in that it involves both discrete and continuous random variables, the position of the polymer being continuous, while the number of monomers in the polymer is discrete. The authors derive an ordinary differential equation for the probability of there being exactly n polymers at a particular time. From this they show how to obtain sample paths for polymer growth and give a brief discussion on the statistics of polymer growth.

Attention is then turned to the modeling of molecular motions, with the first example being the Brownian motion of proteins in aqueous solutions. The (stochastic) Langevin equation is given for the motion of the protein, both with and without an external force acting on the protein. To find a numerical solution of this equation is straightforward, as the authors show. But they caution however that simulation of this solution on a computer is liable to introduce spurious results, and so they derive the Smoluchowski model, a somewhat different way of looking at random motion via the evolution of ensembles of paths. In this formulation the Brownian force is replaced by a diffusion term, and the external force is modeled by a drift term.

The authors then consider the modeling of chemical reactions, which supply the energy to the molecular motors. Because of the time scales involved in these reactions, a correct treatment of them would involve quantum mechanics, but the authors use the Smoluchowski model. The simple reaction model they consider involves a positive ion binding to negatively charged amino acid, and using as reaction coordinate the distance between the ion and the amino acid, study the free energy change as a function of the reaction coordinate.

The numerical simulation of the protein motion is then considered in much greater detail, using an algorithm that preserves detailed balance. This involves converting the problem to a Markov chain and a consideration of the boundary conditions, which the authors do for the case of periodic, reflecting, and absorbing. Euler's method is used to solve the resulting equations for the Markov chain, and after dealing with issues of stability and accuracy, the Crank-Nicolson method is used. The last few sections of the chapter are devoted to the physics of these solutions and the authors give some intuitive feel for the entropic factors and energy balance on a protein motor.

In the last chapter of the book, the considerations in chapter 12 are applied to concrete molecular motors. The first one examined is a model for switching in a bacterial flagellar motor, which involves the protein CheY as a signaling pathway. The binding of CheY to the motor is modeled as a two-state process, with the binding site being either empty or occupied. The resulting set of coupled differential equations for the probabilities is solved for when the concentration of CheY is constant. An expression for the change in free energy is obtained, and the authors give a discussion of the physics in the light of what was done in the last chapter. The switching rate is computed, along with the mean first passage time.

Some other examples of molecular motors are also discussed, including the flashing racket, the polymerization ratchet, and a simplified model of the ion-driven F0 motor of ATP synthase. This latter motor is fascinating, since it describes the electrochemical energy involved in mitochondria for the production of ATP. The authors do a nice job of showing how the techniques of chapter 12 are used to solve this model, and also give an analytical solution for a certain limiting case.

Chapter 4 of this book is the pot of gold: A concrete, detailed description of how the cerebral cortex works. Thinking relies upon an operation which could be called confabulation (in the chapter it is called concensus building). THIS IS NOT REASONING (at least in any classical sense). Yet, because the simple kind of knowledge used (antecedent support probabilities) is exhaustive, concensus building yields excellent conclusions. This cortical theory also shows why AI has failed: reasoning is too difficult and requires too much knowledge of an expensive type. Cortex gets by with a much simpler type of knowledge (which only concerns pairs of object and action attributes, not n-tuples) which, while it is needed in huge quantities, is easy to obtain. An implication of this corticl theory is that we can now proceed to develop successful AI by adopting this cortical design. The theory is illustrated by means of computer thinking experiments that yield compelling results (and which readers can replicate).

The computational Aspects of General Relativity, by J. Frauendiener, is one of many rather interesting papers one would find in this book. This book covers the 2nd Russian-German Advanced Research Workshop which took place in March of 2005. The papers that were presented cover the applications of High Performance Computing (HPC) in research areas such as Fluid Mechanics and computational Physics. As one would expect, the underlying pattern of the types of problems faced by these area are the same as any other, mainly they must be parallelizable, but the effort taken by various researchers to essentially breakup a given problem is truly remarkable.

Particle methods in powder technology, by B. Henrich, et. al., is another paper which the authors use the HPC methods to simulate compaction, sintering and filling of a dispersed powder. The simulation, as one would expect, uses a molecular dynamics method which in turn follows Newton's equations of motions for simulating the time evolution of particles. As one would expect, the sheer number of simulations that must take place is simply staggering, and an HPC is required to get the job done is a reasonable amount of time.

There are a number of other paper that talk about the applications of HPC in Fluid Dynamics or Mechanics, but I was very interested in H. Brunst, et. al. paper, Parallel applications on large scale systems: getting insights, which the authors talk about the "scalability properties on moderns parallel computer architectures". The authors evaluate BenchIT, which is used for evaluation and presentation of performance data. BenchIT is used to compare difference architectures and programming languages against each other to determine their efficiency and performance numbers. BenchIT can be used for simply estimating what the gain would be if for example one programming language is used over the other given certain scenarios. It is a great tool for ROI (return on investments) calculations.

U. Kuster, et. al. paper, Sustaining performance in future vector processors, is simply fascinating! The authors present current performance numbers from a number of platforms, and tie the performance to the software architecture - which is very much true. Teraflops/sec performance is desired, but the authors argue that with current processor architecture, the only way to achieve such performance numbers one much re-architect the way we tackle the problem from the software architecture perspective. The authors then close the loop with possible improvements to the processor architecture that can aid in making the software design a bit simpler. The goal of this paper is show that software and hardware are more tied together than ever, and in order to "go to the next level", both need to work together instead of dispersed components.

I was very surprised at the quality of the papers that are in this text. As an architect that deals with Grid and HPC systems on daily basis, I found this text very informative and educational.

Contacted Springer who manufactured cd in microsoft format incompatible to mac today (26Feb2002).

... These authors are masters. Each author, I find from surfing on the web, has a computer learning lab. The early reading introduces the fact that without the constraints of paper, undergraduates can learn to compose solutions to more realistic problems, eg. golf balls do not go on parabolas but do depend on your irons and crosswinds.

My last read text by the same authors, "Nonlinear Physics with Maple" edition 1 was a really inspiring book full of exercises. A master of solitons, Ablowitz, wrote (with) "Complex Variables" available at [Amazon.com]. Riemann Hilbert problems chapter 7 ought to further flesh out "Nonlinear Physics with Maple" last chapter concerning inverse scattering method in a to me readable manner.

This book has the best index I've ever seen. The index alone is worth the price of the book. I could spend hours going over all those entries. Fabulous.

In this book the author makes a case for musical creativity by machines resulting from a process named "inductive association" which is a more narrowly defined version of free association, or the shedding of deductive reasoning for a more intuitive process. Each chapter begins with a simple principle, one that the author attempts to prove as the chapter proceeds. These principles are followed by illustrative vignettes appropriate to that chapter's focus. Many chapters also contain descriptions of computer programs designed to demonstrate that chapter's focus on the complexities of musical creativity.
Part one provides the history and meaning of creativity. That section ends with a detailed analysis of randomness and how it differs from creativity. There is a summary of computer program types that in some way may model musical creativity.
Part two describes a number of possible models for computationally imitating human creativity. Chapter 4 outlines the basic principles of recombinance and pattern matching, which are the two foundations of the author's work with computers and music. Chapter 5 describes how allusions contribute to musical creativity, and concludes with the description of a program that analyzes music for its references to other music and possible ways in which these references might be interpreted. Chapter 6 explains the role that learning plays in the creative process. Chapter 7 presents some of the ways in which composers build musical expectations and then either fulfill them or surprise listeners. It then discusses musical hierarchy and how computer programs can incorporate the analytical tools necessary to meld hierarchy into their creative processes.
Part 3 presents an inductive-association computational process that can solve problems and produce music creatively. Chapter 9 defines association networks and explains how such networks can respond effectively to input. Chapter 10 applies the principles of association networks to music. Chapter 11 discusses a number of possible combinations of the processes discussed so far, ultimately favoring an integrative model. Chapter 12 presents a number of aesthetical difficulties involved when computationally modeling creativity.
Readers of this book should have experience with music notation, but no knowledge of computer programming is required. However, I think it would have been more difficult to get through this book if I didn't already know AI terminology. All of the example programs are written in Common LISP. This book has a very academic tone which should be familiar to anybody who is familiar with Cope's other works. I recommend it for anyone who has the appropriate background who is interested in whether or not computers "think", and if they do, if they can also "create". Amazon does not show the table of contents so I do that here:
Preface
I Background and Principles 1
1 Definitions
2 Background 35
3 Current Models of Musical Creativity 51
II Experimental Models of Musical Creativity 85
4 Recombinance 87
5 Allusion 125
6 Learning, Inference, and Analogy 177
7 Form and Structure 221
8 Influence 251
III An Integrated Model of Musical Creativity 269
9 Association 271
10 Musical Association 299
11 Integration 325
12 Aesthetics 345
Bibliography
Appendix A: Experiments In Musical Intelligence Final Work List 385
Appendix B: Database Format 391
Appendix C: Ark Endings 393
Appendix D: Listing of Book Programs 397
Appendix E: Virtual Beethoven Symphony No. 10, Second Movement 399
Index

From the table of contents this book has a decided statistical bent but it does this to cover important issues of validation and reliability of models. The sections on discrete and systems dynamics modelling are both thorough and informative. I wondered if the computer programs were afterthoughts rather than necessary but it hardly matters given the book's value when it talks about the concepts and the practical considerations.

Heerman's book is very different from most simulation books: it short. This is achieved by only giving brief explanations of the ideas and methods. It covers most aspects of basic (and not so basic) Molecular Dynamics and Monte Carlo, and since it's so short you can always read it all and then choose what you want to do. It makes simulation look really easy, and includes all the fortran code you need to prove it. Integration methods are explained in a particular way that makes them really easy to use. If you want an overview of the methods and/or are in a hurry read this book. Otherwise go to Allen's or Rapaport's. If you are a beginner I suggest Haile's book, but then come back to this one, you won't regret it.