Pre-trained Text-to-Image Diffusion Models Are Versatile Representation Learners for Control

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gupta, Gunshi, Yadav, Karmesh, Gal, Yarin, Batra, Dhruv, Kira, Zsolt, Lu, Cong, Rudner, Tim G. J.
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917662231101440
author Gupta, Gunshi
Yadav, Karmesh
Gal, Yarin
Batra, Dhruv
Kira, Zsolt
Lu, Cong
Rudner, Tim G. J.
author_facet Gupta, Gunshi
Yadav, Karmesh
Gal, Yarin
Batra, Dhruv
Kira, Zsolt
Lu, Cong
Rudner, Tim G. J.
contents Embodied AI agents require a fine-grained understanding of the physical world mediated through visual and language inputs. Such capabilities are difficult to learn solely from task-specific data. This has led to the emergence of pre-trained vision-language models as a tool for transferring representations learned from internet-scale data to downstream tasks and new domains. However, commonly used contrastively trained representations such as in CLIP have been shown to fail at enabling embodied agents to gain a sufficiently fine-grained scene understanding -- a capability vital for control. To address this shortcoming, we consider representations from pre-trained text-to-image diffusion models, which are explicitly optimized to generate images from text prompts and as such, contain text-conditioned representations that reflect highly fine-grained visuo-spatial information. Using pre-trained text-to-image diffusion models, we construct Stable Control Representations which allow learning downstream control policies that generalize to complex, open-ended environments. We show that policies learned using Stable Control Representations are competitive with state-of-the-art representation learning approaches across a broad range of simulated control settings, encompassing challenging manipulation and navigation tasks. Most notably, we show that Stable Control Representations enable learning policies that exhibit state-of-the-art performance on OVMM, a difficult open-vocabulary navigation benchmark.
format Preprint
id arxiv_https___arxiv_org_abs_2405_05852
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Pre-trained Text-to-Image Diffusion Models Are Versatile Representation Learners for Control
Gupta, Gunshi
Yadav, Karmesh
Gal, Yarin
Batra, Dhruv
Kira, Zsolt
Lu, Cong
Rudner, Tim G. J.
Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
Machine Learning
Robotics
Embodied AI agents require a fine-grained understanding of the physical world mediated through visual and language inputs. Such capabilities are difficult to learn solely from task-specific data. This has led to the emergence of pre-trained vision-language models as a tool for transferring representations learned from internet-scale data to downstream tasks and new domains. However, commonly used contrastively trained representations such as in CLIP have been shown to fail at enabling embodied agents to gain a sufficiently fine-grained scene understanding -- a capability vital for control. To address this shortcoming, we consider representations from pre-trained text-to-image diffusion models, which are explicitly optimized to generate images from text prompts and as such, contain text-conditioned representations that reflect highly fine-grained visuo-spatial information. Using pre-trained text-to-image diffusion models, we construct Stable Control Representations which allow learning downstream control policies that generalize to complex, open-ended environments. We show that policies learned using Stable Control Representations are competitive with state-of-the-art representation learning approaches across a broad range of simulated control settings, encompassing challenging manipulation and navigation tasks. Most notably, we show that Stable Control Representations enable learning policies that exhibit state-of-the-art performance on OVMM, a difficult open-vocabulary navigation benchmark.
title Pre-trained Text-to-Image Diffusion Models Are Versatile Representation Learners for Control
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
Machine Learning
Robotics
url https://arxiv.org/abs/2405.05852