Learning Reward Functions from Scale Feedback
This addresses the problem of improving human-robot interaction for inexperienced users, though it appears incremental as it builds on existing preference learning frameworks.
The paper tackles the problem of robots learning user preferences more efficiently by introducing scale feedback (using a slider) instead of binary choices, which provides more nuanced information. The result is a demonstrated performance benefit in simulations and user studies showing more effective learning in practice.
Today's robots are increasingly interacting with people and need to efficiently learn inexperienced user's preferences. A common framework is to iteratively query the user about which of two presented robot trajectories they prefer. While this minimizes the users effort, a strict choice does not yield any information on how much one trajectory is preferred. We propose scale feedback, where the user utilizes a slider to give more nuanced information. We introduce a probabilistic model on how users would provide feedback and derive a learning framework for the robot. We demonstrate the performance benefit of slider feedback in simulations, and validate our approach in two user studies suggesting that scale feedback enables more effective learning in practice.