Permafrost in the Arctic is thawing faster than ever, according to a new US government report that also found Arctic seawater is war...
domenica 16 agosto 2015
Mass extinction may accelerate robot evolution
A new study using computer simulated robots suggests that mass extinctions may speed up evolution.
In a study published in the journal PLOS One, the computer simulated robots evolved more quickly and efficiently after a virtual mass extinction based on events such as the one that killed off the dinosaurs. The study shows that mass extinction actually speeds up evolution by creating more adaptivity.
“Focused destruction can lead to surprising outcomes,” said co-author Risto Miikkulainen, professor of computer science at The University of Texas at Austin in US. “Sometimes you have to develop something that seems objectively worse in order to develop the tools you need to get better.”
Some evolutionary biologists hypothesize that extinction events actually accelerate evolution by promoting those lineages which evolve the best, meaning ones that can quickly create useful new features and abilities.
The innovation in the latest research was in examining how mass extinction could aid in computational evolution. In computer simulations, scientists connected computer algorithms, or neural networks, to simulated robotic legs with the goal of evolving a robot that could walk smoothly and stably.
Random mutations were introduced through the computational evolution process. They created many different niches so that a wide range of features and abilities would come about. After hundreds of generations, a wide range of robotic behaviors had evolved to fill these niches, although many were not directly useful for walking. Then the researchers randomly killed off the robots in 90 per cent of the niches, mimicking a mass extinction.
After several such cycles of evolution and extinction, they found that the robots that survived were the most evolved and had the greatest potential to produce new behaviors. In some cycles, better solutions than walking were evolved in simulations which contained mass extinctions rather than ones without.
This research could aid in other areas, such as the development of robots that can better overcome obstacles (such as robots searching for survivors in earthquake rubble, exploring Mars or navigating a minefield) and more human-like gaming.