A human uses reinforcements to teach a robot to move a pink ball one foot from a blue ball. The robot acts autonomously, choosing its next action based on its current belief about where the next reinforcement will come.
@passenger1982 I can't seem to link to the paper because of youtube's restrictions, but you can find the paper describing this work by googling "Mark Woodward Harvard", and looking under the Publications section. Please feel free to contact me with questions or comments. A paper extending this work will come out early next year.
@passenger1982 I can't seem to link to the paper because of youtube's restrictions, but you can find the paper describing this work by googling "Mark Woodward Harvard", and looking under the Publications section. Please feel free to contact me with questions or comments. A paper extending this work will come out early next year.
markpwoodward 2 years ago
Very interesting. Do you have more information on this project?
passenger1982 2 years ago