Feature Expansive Reward Learning: Rethinking Human Input

By Andreea Bobu, Marius Wiggert, Claire Jennifer Tomlin, Anca Dragan. 2020

49 citations recorded by the source corpus.

When a person is not satisfied with how a robot performs a task, they can intervene to correct it. Reward learning methods enable the robot to adapt its reward function online based on such human input, but they rely on handcrafted features. When the correction cannot be explained by these features, recent work in deep Inverse Reinforcement Learning (IRL) suggests that the robot could ask for task demonstrations and recover a reward defined over the raw state space. Our insight is that rather than implicitly learning about the missing feature(s) from demonstrations, the robot should instead ask for data that explicitly teaches it about what it is missing. We introduce a new type of human input in which the person guides the robot from states where the feature being taught is highly expressed to states where it is not. We propose an algorithm for learning the feature from the raw state space and integrating it into the reward function. By focusing the human input on the missing feature, our method decreases sample complexity and improves generalization of the learned reward over the above deep IRL baseline. We show this in experiments with a physical 7DOF robot manipulator, as well as in a user study conducted in a simulated environment.

Subfields: Artificial Intelligence, Cognitive Neuroscience, Control and Systems Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Lin Guan
2. Ruohan Zhang
3. Mudit Verma
4. Subbarao Kambhampati
5. Sihang Guo
6. Sarath Sreedharan

In Cognitive Neuroscience

1. Lin Guan
2. Ruohan Zhang

In Control and Systems Engineering

1. Lin Guan
2. Mudit Verma
3. Subbarao Kambhampati
4. Sihang Guo
5. Ruohan Zhang

Citations in the corpus

• Widening the Pipeline in Human-Guided Reinforcement Learning with Explanation and Context-Aware Data Augmentation (2020, background): “EXPAND, thus, can also be useful in aligning objectives of humans in the loop to the agent’s [15, 5].”