Uncertainty-Aware Data Aggregation for Deep Imitation Learning

Estimating statistical uncertainties allows autonomous agents to communicate their confidence during task execution and is important for applications in safety-critical domains such as autonomous driving. In this work, we present the uncertainty-aware imitation learning (UAIL) algorithm for improving end-to-end control systems via data aggregation. UAIL applies Monte Carlo Dropout to estimate uncertainty in the control output of end-to-end systems, using states where it is uncertain to selectively acquire new training data. In contrast to prior data aggregation algorithms that force human experts to visit sub-optimal states at random, UAIL can anticipate its own mistakes and switch control to the expert in order to prevent visiting a series of sub-optimal states. Our experimental results from simulated driving tasks demonstrate that our proposed uncertainty estimation method can be leveraged to reliably predict infractions. Our analysis shows that UAIL outperforms existing data aggregation algorithms on a series of benchmark tasks.

Authors

• Yuchen Cui
• David Isele
• Scott Niekum
• Kikuo Fujimura

Keywords

• Uncertainty
• Data aggregation
• Task analysis
• Switches
• Estimation
• Data models
• Aggregate Data
• Imitation Learning
• Training Data
• Control System
• Learning Algorithms
• Uncertainty Estimation
• Task Execution
• Human Experts
• Suboptimal Conditions
• Switching Control
• Series Of States
• Experts In Order
• Receiver Operating Characteristic Curve
• Deep Network
• Collection Process
• Active Learning
• Continuous-time
• User Study
• Random Noise
• Discrete Distribution
• Learning Agent
• Model Predictive Control
• Steering Angle
• Human Effort
• Policy Learning
• Correct Behavior
• Bad Conditions
• Prediction Uncertainty
• Adversary Model
• Output Samples