Active Preference-Based Gaussian Process Regression for Reward Learning
This addresses the problem of data inefficiency and inflexibility in reward learning for robotics, offering a more practical alternative to demonstration-based methods, though it is incremental in improving preference-based frameworks.
The paper tackles the challenge of designing reward functions in AI and robotics by proposing a preference-based learning approach that uses Gaussian Process regression to learn from human comparisons between trajectories, resulting in efficient learning of expressive reward functions as shown in simulations and a user study.
Designing reward functions is a challenging problem in AI and robotics. Humans usually have a difficult time directly specifying all the desirable behaviors that a robot needs to optimize. One common approach is to learn reward functions from collected expert demonstrations. However, learning reward functions from demonstrations introduces many challenges: some methods require highly structured models, e.g. reward functions that are linear in some predefined set of features, while others adopt less structured reward functions that on the other hand require tremendous amount of data. In addition, humans tend to have a difficult time providing demonstrations on robots with high degrees of freedom, or even quantifying reward values for given demonstrations. To address these challenges, we present a preference-based learning approach, where as an alternative, the human feedback is only in the form of comparisons between trajectories. Furthermore, we do not assume highly constrained structures on the reward function. Instead, we model the reward function using a Gaussian Process (GP) and propose a mathematical formulation to actively find a GP using only human preferences. Our approach enables us to tackle both inflexibility and data-inefficiency problems within a preference-based learning framework. Our results in simulations and a user study suggest that our approach can efficiently learn expressive reward functions for robotics tasks.