麻豆淫院

(麻豆淫院Org.com) -- Robots typically acquire new capacities by imitation. Now, EPFL scientists are doing the inverse -- developing machines that can learn more rapidly and outperform humans by starting from failed or inaccurate demonstrations.

A robot, unblinking, impassive, observes. Its instructor wants it to learn how to put a balloon in a basket 20 meters away. As the researcher demonstrates this task, which is difficult for a human to accomplish, she systematically misses the basket. Isn鈥檛 the scientist just wasting her time?

Typically looked at simply as useless mistakes, failed demonstrations can, on the contrary, be opportunities to learn better, claim scientists from EPFL鈥檚 Learning Algorithms and Systems Laboratory (LASA). Their unusual point of view has led to the development of novel algorithms.

鈥淲e inversed the principle, generally accepted in robotics, of acquisition by imitation, and considered cases in which humans are inaccurate in certain tasks,鈥� explains professor Aude Billard, head of LASA. 鈥淭his approach allows the robot to go further, to learn more quickly and above all, outperform the human,鈥� notes postdoctoral researcher Dan Grollman, who was recently awarded a 鈥淏est Paper Award鈥� for an article on the subject presented at the International Conference on Robotics and Automation (ICRA), in Shanghai.

Grollman based his work on what he calls the 鈥淒onut as I do鈥� theory. He developed an algorithm that tells the robot not to reproduce a demonstrator鈥檚 inaccurate gesture. The machine will use this input to avoid repeating the mistake and to search for alternative solutions. Thus the choice of the term 鈥渄onut鈥� 鈥� a play on the words 鈥渄o not鈥� and 鈥渄onut.鈥� The hole in the middle is the incorrect gesture, which must be excluded, and the surrounding dough represents the field of potential solutions to explore.

鈥淲e were inspired by the way in which humans learn,鈥� explains Billard. 鈥淐hildren often progress by making mistakes or by observing others鈥� mistakes and assimilating the fact that they must not reproduce them.鈥�

This way of learning things is a 鈥渞eal step forward,鈥� says Grollman, who adds that, after all, 鈥渋sn鈥檛 the real goal to make robots that can do things we can鈥檛?鈥�