Visual Adversarial Imitation Learning using Variational Models

Abstract

Reward function specification, which requires considerable human effort and iteration, remains a major impediment for learning behaviors through deep reinforcement learning. In contrast, providing visual demonstrations of desired behaviors often presents an easier and more natural way to teach agents. We consider a setting where an agent is provided a fixed dataset of visual demonstrations illustrating how to perform a task, and must learn to solve the task using the provided demonstrations and unsupervised environment interactions. This setting presents a number of challenges including representation learning for visual observations, sample complexity due to high dimensional spaces, and learning instability due to the lack of a fixed reward or learning signal. Towards addressing these challenges, we develop a variational model-based adversarial imitation learning (V-MAIL) algorithm. The model-based approach provides a strong signal for representation learning, enables sample efficiency, and improves the stability of adversarial training by enabling onpolicy learning. Through experiments involving several vision-based locomotion and manipulation tasks, we find that V-MAIL learns successful visuomotor policies in a sample-efficient manner, has better stability compared to prior work, and also achieves higher asymptotic performance. We further find that by transferring the learned models, V-MAIL can learn new tasks from visual demonstrations without any additional environment interactions. All results including videos can be found online at https://sites.google.com/view/variational-mail.

Off-policy Imitation Learning from Visual Inputs

“…Our work bears some resemblance to the two concurrent works Rafailov et al (2021); Anonymous (2022). Compared to Rafailov et al (2021), we solve ILfVI from a model-free perspective, which improves the performance with merely off-policy samples instead of a learned model and even works for visual observations.…”

“…Torabi et al (2018b) conduct experiments of GAIfO with visual observations, showing that GAIfO only achieves about half of the expert-level performance in no-trivial environments. Two concurrent works (Rafailov et al, 2021;Anonymous, 2022) give further insights into ILfVI. Rafailov et al (2021) solve ILfVI from a model-based perspective, whose algorithm V-MAIL first learns a world model and then updates the discriminator with on-policy samples from the learned model.…”

“…Two concurrent works (Rafailov et al, 2021;Anonymous, 2022) give further insights into ILfVI. Rafailov et al (2021) solve ILfVI from a model-based perspective, whose algorithm V-MAIL first learns a world model and then updates the discriminator with on-policy samples from the learned model. Anonymous (2022) study IL from visual observations with a model-free scheme.…”

“…For example, Laskin et al (2020) illustrate that general data augmentation methods enable agents to achieve excellent performance in RL from visual inputs with extensive studies. In ILfVI, data augmentation also demonstrates remarkable performance gains (Young et al, 2020;Rafailov et al, 2021).…”

“…The thought to enable agents to learn like intelligent creatures lead to a spectrum of IL, imitation learning from visual inputs (or images, or pixels) (ILfVI). This spectrum of IL is also referred to as visual imitation learning (VIL) (Young et al, 2020;Rafailov et al, 2021). Here, we denote it as ILfVI to emphasize that visual inputs can be further classified.…”