Artificial Natural Selection?
How to train a neural network for self-driving vehicles? The team of Waymo has taken a cue from biology to speed things up. On top of that, reinforcement learning is applied to steer the ‘evolution’ in the desired direction:
“In these experiments, we were able to increase diversity by creating sub-populations called “niches,” where neural nets were only allowed to compete within their own sub-groups–similar to how species evolve when isolated on islands. We also tried to directly reward diversity through a technique called “fitness sharing,” where we measure the difference between members of the population and give more unique neural nets an edge in the competition. Greater diversity allows PBT to explore a larger hyperparameter space.” - How evolutionary selection can train more capable self-driving cars - Deepmind
Next to Darwin’s natural selection, we now have artificial natural selection as well?