What type of study is this?

This is a Quantitative Study study.

October 5, 2025Open Access

TubeDAgger: Reducing the Number of Expert Interventions with Stochastic Reach-Tubes

Key Points

This approach reduces the number of expert interventions needed during the training of novice policies in interactive imitation learning.
Using stochastic reach-tubes as a decision-making tool led to fewer expert interventions, outperforming related methods.
Incorporating a new method avoids the need for threshold tuning, making the model easier to apply across various environments.
The findings highlight the effectiveness of using reach-tubes over traditional doubt classification models for expert control.

Abstract

Interactive Imitation Learning deals with training a novice policy from expert demonstrations in an online fashion. The established DAgger algorithm trains a robust novice policy by alternating between interacting with the environment and retraining of the network. Many variants thereof exist, that differ in the method of discerning whether to allow the novice to act or return control to the expert. We propose the use of stochastic reachtubes - common in verification of dynamical systems - as a novel method for estimating the necessity of expert intervention. Our approach does not require fine-tuning of decision thresholds per environment and effectively reduces the number of expert interventions, especially when compared with related approaches that make use of a doubt classification model.

Read Full Paperexternally

Ask AI

Helpful

Bookmark

View Full Paper