Using Machine Teaching to Investigate Human Assumptions when Teaching Reinforcement Learners
arXiv:2009.02476
Abstract
Successful teaching requires an assumption of how the learner learns - how the learner uses experiences from the world to update their internal states. We investigate what expectations people have about a learner when they teach them in an online manner using rewards and punishment. We focus on a common reinforcement learning method, Q-learning, and examine what assumptions people have using a behavioral experiment. To do so, we first establish a normative standard, by formulating the problem as a machine teaching optimization problem. To solve the machine teaching optimization problem, we use a deep learning approximation method which simulates learners in the environment and learns to predict how feedback affects the learner's internal states. What do people assume about a learner's learning and discount rates when they teach them an idealized exploration-exploitation task? In a behavioral experiment, we find that people can teach the task to Q-learners in a relatively efficient and effective manner when the learner uses a small value for its discounting rate and a large value for its learning rate. However, they still are suboptimal. We also find that providing people with real-time updates of how possible feedback would affect the Q-learner's internal states weakly helps them teach. Our results reveal how people teach using evaluative feedback and provide guidance for how engineers should design machine agents in a manner that is intuitive for people.
arXiv comment21 pages, 4 figures
References in corpus (14)
- Addressing Function Approximation Error in Actor-Critic Methods
- Interactive Learning from Policy-Dependent Human Feedback
- An Overview of Machine Teaching
- Iterative Machine Teaching
- Data Poisoning Attacks against Online Learning
- Adversarial Attacks on Stochastic Bandits
- Policy Poisoning in Batch Reinforcement Learning and Control
- Policy Teaching via Environment Poisoning: Training-time Adversarial Attacks against Reinforcement Learning
- Teaching Inverse Reinforcement Learners via Features and Demonstrations
- Adaptive Reward-Poisoning Attacks against Reinforcement Learning
- Preference-Based Batch and Sequential Teaching: Towards a Unified View of Models
- Online Data Poisoning Attack
- Machine Teaching of Active Sequential Learners
- An Optimal Control Approach to Sequential Machine Teaching