The findings can have important implications for the future of AI and robotics research.
Evolution is hard to simulate
40% off TNW Conference!
Mosquitos have thermal vision to spot body heat.
Bats have wings to fly and an echolocation apparatus to navigate dark places.
Sea turtles have flippers to swim and a magnetic field detector system to travel very long distances.
Interestingly, all these species descended from the first lifeform that appeared on Earth several billion years ago.
Studying the evolution of life and intelligence is interesting.
Butreplicating it is extremely difficult.
It would need a lot of parallel and sequential trial-and-error cycles.
AI researchers use several shortcuts and predesigned features to overcome some of these challenges.
Yet another approach is to train different AI subsystems separately (vision, locomotion, language, etc.)
and then tack them on together in a final AI or robotic system.
Their framework is called Deep Evolutionary Reinforcement Learning.
In DERL each agent usesdeep reinforcement learningto acquire the skills required to maximize its goals during its lifetime.
None of the learned parameters are passed on across generations.
Each agent in the environment is composed of a genotype that defines its limbs and joints.
Each agent is trained with reinforcement learning to maximize rewards in various environments.
Agents whose physical structure are better suited for traversing terrain learn faster to use their limbs for moving around.
The flat terrain puts the least selection pressure on the agents morphology.
The MVT variant has the added challenge of requiring the agents to manipulate objects to achieve their goals.
it’s possible for you to read the original articlehere.
