Evolving Open Complexity
This addresses a bottleneck in evolutionary computation for AI, offering an incremental improvement to enhance the efficiency of evolving complex programs.
The paper investigates the challenge of evolving large, complex programs through genetic programming, finding that many genetic changes are silent and do not affect outputs, which slows evolution. It proposes adopting an open architecture with mutation sites close to the environment to enable measurable impacts and faster evolution.
Information theoretic analysis of large evolved programs produced by running genetic programming for up to a million generations has shown even functions as smooth and well behaved as floating point addition and multiplication loose entropy and consequently are robust and fail to propagate disruption to their outputs. This means, while dependent upon fitness tests, many genetic changes deep within trees are silent. For evolution to proceed at reasonable rate it must be possible to measure the impact of most code changes, yet in large trees most crossover sites are distant from the root node. We suggest to evolve very large very complex programs, it will be necessary to adopt an open architecture where most mutation sites are within 10 to 100 levels of the organism's environment.