Why walk when you can flop?
In one example, a simulated robot was supposed to evolve to travel as quickly as possible. But rather than evolve legs, it simply assembled itself into a tall tower, then fell over. Some of these robots even learned to turn their falling motion into a somersault, adding extra distance.

Blog by Janelle Shane: https://lnkd.in/dQnCVa9

Original paper: https://lnkd.in/dt63hJR

#algorithm #artificialintelligence #machinelearning #reinforcementlearning #technology

----------
@machinelearning_tuts

How do you go from self-play to the real world? : Transfer learning

NeurIPS 2017 Meta Learning Symposium: https://lnkd.in/e7MdpPc

A new research problem has therefore emerged: How can the complexity, i.e. the design, components, and hyperparameters, be configured automatically so that these systems perform as well as possible? This is the problem of metalearning. Several approaches have emerged, including those based on Bayesian optimization, gradient descent, reinforcement learning, and evolutionary computation.

#artificialintelligence #deeplearning #metalearning #reinforcementlearning
----------
@machinelearning_tuts