3 ms·
Yes, random mutations. Random mutations that cause cancer kill an organism and random mutations that make them more attractive to mates make them more likely to
by heurist 8y ago
Yes, random mutations. Random mutations that cause cancer kill an organism and random mutations that make them more attractive to mates make them more likely to pass that mutations to offspring, who pass it to their offspring, and so on. This is all basic evolutionary theory...
- p1esk 8y agoYes, I'm questioning the "basic evolutionary theory" (and it appears that Dyson does too), because all efforts to test it in software have failed so far. Which is especially troubling given that processing power has been increasing exponentially for decades since people started experimenting with genetic algorithms and artificial life simulations.
- heurist 8y ago> all efforts to test it in software have failed so far What does "failed" mean? Evolution is a universal concept with infinite potential implementations. Trying to measure "success" means nothing unless you have some particular goal in mind, and as soon as you add a goal you severely restrict the pool of potential "successful" implementations, which makes success harder to find. Even then, you can look to processes like stochastic gradient descent as a "successful" use. It goes a lot deeper than that but as someone who has spent many years studying evolutionary biology, computer science, information theory, cognitive science, socioeconomics, and complex adaptive systems I have no idea where your comment is coming from so I won't elaborate here.