Tetris AI Generation Using Nelder-Mead and Genetic Algorithms
نویسنده
چکیده
In this paper, we discuss the training of a tetris playing AI using a combination of Nelder-Mead optimization and genetic algorithms. The core AI consisted of a variable depth look-ahead player optimizing a set of 16 features, built on the core Java code provided by the TA. A set of random scores were optimized using MATLAB’s implementation of Nelder-Mead, fed through a genetic algorithm, then passed through Nelder-Mead again. When run at depth one look-ahead (testing the placement of two pieces), the resulting program averaged 160,000 lines cleared, and cleared over 430,000 lines in the best of 16 trials.
منابع مشابه
A HYBRID MODIFIED GENETIC-NELDER MEAD SIMPLEX ALGORITHM FOR LARGE-SCALE TRUSS OPTIMIZATION
In this paper a hybrid algorithm based on exploration power of the Genetic algorithms and exploitation capability of Nelder Mead simplex is presented for global optimization of multi-variable functions. Some modifications are imposed on genetic algorithm to improve its capability and efficiency while being hybridized with Simplex method. Benchmark test examples of structural optimization with a...
متن کاملA Framework for Derivative Free Algorithm Hybridization
Column generation is a basic tool for the solution of largescale mathematical programming problems. We present a class of column generation algorithms in which the columns are generated by derivative free algorithms, like population-based algorithms. This class can be viewed as a framework to define hybridization of free derivative algorithms. This framework has been illustrated in this article...
متن کاملNelder-Mead Evolutionary Hybrid Algorithms
Real world optimization problems are often too complex to be solved through analytic means. Evolutionary algorithms are a class of algorithms that borrow paradigms from nature to address them. These are stochastic methods of optimization that maintain a population of individual solutions, which correspond to points in the search space of the problem. These algorithms have been immensely popular...
متن کاملHybrid Simulated Annealing and Nelder-mead Algorithm for Solving Large-scale Global Optimization Problems
This paper presents a new algorithm for solving large scale global optimization problems based on hybridization of simulated annealing and Nelder-Mead algorithm. The new algorithm is called simulated Nelder-Mead algorithm with random variables updating (SNMRVU). SNMRVU starts with an initial solution, which is generated randomly and then the solution is divided into partitions. The neighborhood...
متن کاملFrom Evolutionary Operation to Parallel Direct Search: Pattern Search Algorithms for Numerical Optimization
G.E.P. Box’s seminal suggestions for Evolutionary Operation led other statisticians to propose algorithms for numerical optimization that rely exclusively on the direct comparison of function values. These contributions culminated in the development of the widely used simplex algorithm of Nelder and Mead. Recent examination of these popular methods by the numerical optimization community has pr...
متن کامل