Kotanchek Mark - Genetic Programming Theory and Practice XI
Here you can read online Kotanchek Mark - Genetic Programming Theory and Practice XI full text of the book (entire story) in english for free. Download pdf and epub, get meaning, cover and reviews about this ebook. City: New York;NY, year: 2014, publisher: Springer New York, genre: Home and family. Description of the work, (preface) as well as reviews are available. Best literature library LitArk.com created for fans of good reading and offers a wide selection of genres:
Romance novel
Science fiction
Adventure
Detective
Science
History
Home and family
Prose
Art
Politics
Computer
Non-fiction
Religion
Business
Children
Humor
Choose a favorite category and find really read worthwhile books. Enjoy immersion in the world of imagination, feel the emotions of the characters or learn something new for yourself, make an fascinating discovery.
- Book:Genetic Programming Theory and Practice XI
- Author:
- Publisher:Springer New York
- Genre:
- Year:2014
- City:New York;NY
- Rating:3 / 5
- Favourites:Add to favourites
- Your mark:
Genetic Programming Theory and Practice XI: summary, description and annotation
We offer to read an annotation, description, summary or preface (depends on what the author of the book "Genetic Programming Theory and Practice XI" wrote himself). If you haven't found the necessary information about the book — write in the comments, we will try to find it.
These contributions, written by the foremost international researchers and practitioners of Genetic Programming (GP), explore the synergy between theoretical and empirical results on real-world problems, producing a comprehensive view of the state of the art in GP. Topics in this volume include: evolutionary constraints, relaxation of selection mechanisms, diversity preservation strategies, flexing fitness evaluation, evolution in dynamic environments, multi-objective and multi-modal selection, foundations of evolvability, evolvable and adaptive evolutionary operators, foundation of injecting expert knowledge in evolutionary search, analysis of problem difficulty and required GP algorithm complexity, foundations in running GP on the cloud - communication, cooperation, flexible implementation, and ensemble methods. Additional focal points for GP symbolic regression are: (1) The need to guarantee convergence to solutions in the function discovery mode; (2) Issues on model validation; (3) The need for model analysis workflows for insight generation based on generated GP solutions - model exploration, visualization, variable selection, dimensionality analysis; (4) Issues in combining different types of data. Readers will discover large-scale, real-world applications of GP to a variety of problem domains via in-depth presentations of the latest and most significant results.;Extreme Accuracy in Symbolic Regression -- Exploring Interestingness in a Computational Evolution System for the Genome-Wide Genetic Analysis of Alzheimers Disease -- Optimizing a Cloud Contract Portfolio using Genetic Programming-based Load Models -- Maintenance of a Long Running Distributed Genetic Programming System for Solving Problems Requiring Big Data -- Grounded Simulation: Using Simulated Evolution to Guide Embodied Evolution -- Applying Genetic Programming in Business Forecasting -- Explaining Unemployment Rates with Symbolic Regression -- Uniform Linear Transformation with Repair and Alternation in Genetic Programming -- A Deterministic and Symbolic Regression Hybrid Applied to Resting-State fMRI Data -- Gaining Deeper Insights in Symbolic Regression -- Geometric Semantic Genetic Programming for Real Life Applications -- Evaluation of Parameter Contribution to Neural Network Size and Fitness in ATHENA for Genetic Analysis.
Kotanchek Mark: author's other books
Who wrote Genetic Programming Theory and Practice XI? Find out the surname, the name of the author of the book and a list of all author's works by series.
Kotanchek Mark
John Foster
Riolo Rick
August John Hoffman
Michael Doebeli
Manoj Tiwari
Larry Samuelson
Genetic Programming Theory and Practice XI — read online for free the complete book (whole text) full work
Below is the text of the book, divided by pages. System saving the place of the last page read, allows you to conveniently read the book "Genetic Programming Theory and Practice XI" online for free, without having to search again every time where you left off. Put a bookmark, and you can go to the page where you finished reading at any time.
Light
Font size:
↓
↑
Reset
Interval:
↓
↑
Bookmark:
Make
Rick Riolo , Jason H. Moore and Mark Kotanchek (eds.) Genetic and Evolutionary Computation Genetic Programming Theory and Practice XI 2014 10.1007/978-1-4939-0375-7_1
Springer Science+Business Media New York 2014
1. Extreme Accuracy in Symbolic Regression
Michael F. Korns 1
(1)
Analytic Research Foundation, 98 Perea Street, Makati, 1229, Manila, Philippines
Michael F. Korns
Email:
Abstract
Although recent advances in symbolic regression (SR) have promoted the field into the early stages of commercial exploitation, the poor accuracy of SR is still plaguing even the most advanced commercial packages today. Users expect to have the correct formula returned, especially in cases with zero noise and only one basis function with minimally complex grammar depth. Poor accuracy is a hindrance to greater academic and industrial acceptance of SR tools.
In a previous paper, the poor accuracy of Symbolic Regression was explored, and several classes of test formulas, which prove intractable for SR, were examined. An understanding of why these test problems prove intractable was developed. In another paper a baseline Symbolic Regression algorithm was developed with specific techniques for optimizing embedded real numbers constants. These previous steps have placed us in a position to make an attempt at vanquishing the SR accuracy problem.
In this chapter we develop a complex algorithm for modern symbolic regression which is extremely accurate for a large class of Symbolic Regression problems. The class of problems, on which SR is extremely accurate, is described in detail. A definition of extreme accuracy is provided, and an informal argument of extreme SR accuracy is outlined in this chapter. Given the critical importance of accuracy in SR, it is our suspicion that in the future all commercial Symbolic Regression packages will use this algorithm or a substitute for this algorithm.
Keywords
Abstract expression grammars Grammar template genetic programming Genetic algorithms Particle swarm Symbolic regressionIntroduction
The discipline of Symbolic Regression (SR) has matured significantly in the last few years. There is at least one commercial package on the market for several years ( ).
Yet, despite the increasing sophistication of commercial SR packages, there have been serious issues with SR accuracy even on simple problems (Korns ). Clearly the perception of SR as a must use tool for important problems or as an interesting heurism for shedding light on some problems, will be greatly affected by the demonstrable accuracy of available SR algorithms and tools. The depth and breadth of SR adoption in industry and academia will be greatest if a very high level of accuracy can be demonstrated for SR algorithms.
In Korns ().
In this chapter we enhance the previous baseline with a complex multi-island algorithm for modern symbolic regression which is extremely accurate for a large class of Symbolic Regression problems. The class of problems, on which SR is extremely accurate, is described in detail. A definition of extreme accuracy is provided, and an informal argument of extreme SR accuracy is outlined in this chapter.
Before continuing with the details of our extreme accuracy algorithm, we proceed with a basic introduction to general nonlinear regression. Nonlinear regression is the mathematical problem which Symbolic Regression aspires to solve. The canonical generalization of nonlinear regression is the class of Generalized Linear Models (GLMs) as described in Nelder and Wedderburn (). A GLM is a linear combination of I basis functions B i ; i=0,1, I, a dependent variable y, and an independent data point with M features x=0, x1, x2, , x M 1>: such that
- ( E1 ) y=(x)=c0+c i B i (x)+ err
As a broad generalization, GLMs can represent any possible nonlinear formula. However the format of the GLM makes it amenable to existing linear regression theory and tools since the GLM model is linear on each of the basis functions B i . For a given vector of dependent variables, Y, and a vector of independent data points, X, symbolic regression will search for a set of basis functions and coefficients which minimize err . In Koza () the basis functions selected by symbolic regression will be formulas as in the following examples:
- ( E2 ) B0=x3
- ( E3 ) B1=x1+x4
- ( E4 ) B2=sqrt(x2)/tan(x5/4.56)
- ( E5 ) B3=tanh(cos(x2.2)cube(x5+abs(x1)))
If we are minimizing the normalized least squared error, NLSE (Korns ) the genome and the individual are the same Lisp s-expression which is usually illustrated as a tree. Of course the tree-view of an s-expression is a visual aid, since a Lisp s-expression is normally a list which is a special Lisp data structure. Without altering or restricting basic tree-based GP in any way, we can view the individuals not as trees but instead as s-expressions such as this depth 2 binary tree s-exp: (/ (+ x 2 3.45) ( x 0 x 2 )) , or this depth 2 irregular tree s-exp: (/ (+ x 4 3.45) 2.0) .
In basic GP, applied to symbolic regression, the non-terminal nodes are all operators (implemented as Lisp function calls), and the terminal nodes are always either real number constants or features. The maximum depth of a GP individual is limited by the available computational resources; but, it is standard practice to limit the maximum depth of a GP individual to some manageable limit at the start of a symbolic regression run.
Given any selected maximum depth k, it is an easy process to construct a maximal binary tree s-expression U k , which can be produced by the GP system without violating the selected maximum depth limit. As long as we are reminded that each f represents a function node while each t represents a terminal node, the construction algorithm is simple and recursive as follows.
- ( U 0): t
- ( U 1): (f t t)
- ( U 2): (f (f t t) (f t t))
- ( U 3): (f (f (f t t) (f t t)) (f (f t t) (f t t)))
- ( U k): (f U k 1 U k 1)
The basic GP symbolic regression system (Koza ) contains a set of functions F, and a set of terminals T. If we let tT, and fF 
, where (a, b) = (a) =a , then any basis function produced by the basic GP system will be represented by at least one element of U k . Adding the function allows U k to express all possible basis functions generated by the basic GP system to a depth of k . Note to the reader , the function performs the job of a pass-through function. The function allows a fixed-maximal-depth expression in U k to express trees of varying depth, such as might be produced from a GP system. For instance, the varying depth GP expression x 2 +(x 3 x 5 ) = (x 2 , 0.0)+(x 3 x 5 ) = + ((x 2 , 0.0) (x 3 x 5 )) which is a fixed-maximal-depth expression in U2.
, where (a, b) = (a) =a , then any basis function produced by the basic GP system will be represented by at least one element of U k . Adding the function allows U k to express all possible basis functions generated by the basic GP system to a depth of k . Note to the reader , the function performs the job of a pass-through function. The function allows a fixed-maximal-depth expression in U k to express trees of varying depth, such as might be produced from a GP system. For instance, the varying depth GP expression x 2 +(x 3 x 5 ) = (x 2 , 0.0)+(x 3 x 5 ) = + ((x 2 , 0.0) (x 3 x 5 )) which is a fixed-maximal-depth expression in U2.In addition to the special pass through function , in our system we also make additional slight alterations to improve coverage, reduce unwanted errors, and restrict results from wandering into the complex number range. All unary functions, such as cos , are extended to ignore any extra arguments so that, for all unary functions, cos(a,b) =cos(a) . The sqroot and ln functions are extended for negative arguments so that sqroot(a) =sqroot(abs(a)) and ln(a) =ln(abs(a)) .
Given this formalism of the search space, it is easy to compute the size of the search space, and it is easy to see that the search space is huge even for rather simple basis functions. For our use in this chapter the function set will be the following functions: F= ( 
Light
Font size:
↓
↑
Reset
Interval:
↓
↑
Bookmark:
Make
Similar books «Genetic Programming Theory and Practice XI»
Look at similar books to Genetic Programming Theory and Practice XI. We have selected literature similar in name and meaning in the hope of providing readers with more options to find new, interesting, not yet read works.
John Foster
Riolo Rick
August John Hoffman
Michael Doebeli
Manoj Tiwari
Larry Samuelson
Reviews about «Genetic Programming Theory and Practice XI»
Discussion, reviews of the book Genetic Programming Theory and Practice XI and just readers' own opinions. Leave your comments, write what you think about the work, its meaning or the main characters. Specify what exactly you liked and what you didn't like, and why you think so.