New ‘deep learning’ technique enables robot mastery of skills via trial and error UC Berkeley researchers have developed algorithms that enable robots to learn motor tasks through trial and error using a process that more closely approximates the way humans learn, marking a major milestone in the field of artificial intelligence. By Sarah Yang This team of UC Berkeley researchers has developed alg
![New 'deep learning' technique enables robot mastery of skills via trial and error - Berkeley News](https://cdn-ak-scissors.b.st-hatena.com/image/square/d0dad759b5331ea0a0be45062992c98ec8a4e318/height=288;version=1;width=512/https%3A%2F%2Fnews.berkeley.edu%2Fwp-content%2Fuploads%2F2015%2F05%2Frobot-team750.jpg)