Filters








3 Hits in 2.8 sec

Fast Generation of Big Random Binary Trees [article]

William B. Langdon
2020 arXiv   pre-print
random_tree() is a linear time and space C++ implementation able to create trees of up to a billion nodes for genetic programming and genetic improvement experiments.  ...  Acknowledgements This work was inspired by conversations at Dagstuhl Seminar 18052 on Genetic Improvement of Software.  ...  Funded by EPSRC grant EP/M025853/1. random tree code (intended to be used as part of GPquick) can be found via http://www.cs.ucl. ac.uk/staff/W.Langdon/ftp/gp-code/rand_tree.cc_r1.43  ... 
arXiv:2001.04505v1 fatcat:q7rq4dgthbguhkdjgunbbqq2yq

Faster Genetic Programming GPquick via multicore and Advanced Vector Extensions [article]

W. B. Langdon, W. Banzhaf
2019 arXiv   pre-print
We evolve floating point Sextic polynomial populations of genetic programming binary trees for up to a million generations.  ...  To support unbounded Long-Term Evolution Experiment LTEE GP we use both SIMD parallel AVX 512 bit instructions and 48 threads to yield performance of up to 139 billion GP operations per second, 139 giga  ...  Acknowledgements This work was inspired by conversations at Dagstuhl Seminar 18052 on Genetic Improvement of Software.  ... 
arXiv:1902.09215v1 fatcat:5pgwhfc5znfx5muly2u7jzcmp4

Continuous Long-Term Evolution of Genetic Programming

William B. Langdon, Wolfgang Banzhaf
2019 The 2019 Conference on Artificial Life   unpublished
We evolve floating point Sextic polynomial populations of genetic programming binary trees for up to a million generations. Programs with almost 400 000 000 instructions are created by crossover.  ...  To support unbounded Long-Term Evolution Experiment LTEE GP we use both SIMD parallel AVX 512 bit instructions and 48 threads to yield performance of up to 149 billion GP operations per second, 149 giga  ...  Acknowledgements This work was inspired by conversations at Dagstuhl Seminar 18052 on Genetic Improvement of Software.  ... 
doi:10.1162/isal_a_00191 fatcat:rz7uh7mlp5eqhghd6tlrt7rr4a