A Trillion Genetic Programming Instructions per Second

TL;DR

This paper demonstrates that a high-performance computer can interpret over a trillion genetic programming instructions per second, emphasizing parallel computing and information theory insights to optimize evolutionary computation processes.

Contribution

It introduces methods to achieve unprecedented instruction throughput in genetic programming by leveraging parallelism, bandwidth management, and insights from information theory.

Findings

Achieved 1.10337 trillion GPop/s on a 3.0 GHz 16-core AVX512 system.

Highlighting the importance of avoiding deep nesting to prevent evolution stagnation.

Proposes incremental evaluation and FDP techniques to reduce fitness evaluation time.

Abstract

We summarise how a 3.0 GHz 16 core AVX512 computer can interpret the equivalent of up to on average 1103370000000 GPop/s. Citations to existing publications are given. Implementation stress is placed on both parallel computing, bandwidth limits and avoiding repeated calculation. Information theory suggests in digital computing, failed disruption propagation gives huge speed ups as FDP and incremental evaluation can be used to reduce fitness evaluation time in phenotypically converged populations. Conversely FDP may be responsible for evolution stagnation. So the wider Evolutionary Computing, Artificial Life, Unconventional Computing and Software Engineering community may need to avoid deep nesting.