Chelsea Finn：机器人越可靠，DAgger 效率越低
As robots become more reliable, DAgger becomes far less efficient.
随着机器人变得更可靠，DAgger 的效率会大幅下降。 例如 90% 的成功率意味着人 >90% 的时间都在被动地看着机器人。 关键思路：失败来自相似的初始状态，我们可以把学习聚焦在这些状态上！
查看原帖I've been thinking a lot about making robot learning scalable. To me, that means learning on the job, and using the human help still needed efficiently. In Mulligan ⛳️, we focus each data-collection round on the initial states where the robot fails. This happens in deployment, with the robot doing the task and a person stepping in when needed. On 3 real…
展开完整原文
I've been thinking a lot about making robot learning scalable. To me, that means learning on the job, and using the human help still needed efficiently. In Mulligan ⛳️, we focus each data-collection round on the initial states where the robot fails. This happens in deployment, with the robot doing the task and a person stepping in when needed. On 3 real tasks and 2,550 blind evals, it improves on uniform collection at the same budget.
正文
当前显示原文，中文全文译文尚未就绪。
As robots become more reliable, DAgger becomes far less efficient.
eg 90% success means the person spends >90% of the time passively watching the robot
Key idea: failures come from similar initial states & we can focus learning on these states!