R3M: A Universal Visual Representation for Robot Manipulation

# 🌟 New model addition

## Model description
We pre-train a visual representation using the Ego4D human video dataset using a combination of time-contrastive learning, video-language alignment,and an L1 penalty to encourage sparse and compact representations. The resulting representation, R3M, can be used as a frozen perception module for downstream policy learning. Across a suite of 12 simulated robot manipulation tasks, we find that R3M improves task success by over 20% compared to training from scratch and by over 10% compared to state-of-the-art visual representations like CLIP and MoCo. Furthermore, R3M enables a Franka Emika Panda arm to learn a range of manipulation tasks in a real, cluttered apartment given just 20 demonstrations.

## Open source status

* [x] the model implementation is available:(https://github.com/facebookresearch/r3m)
* [x] the model weights are available: https://github.com/facebookresearch/r3m/blob/main/r3m/example.py
* [x] who are the authors: @suraj-nair-1

Start with the linked R3M implementation and weights in r3m/example.py, then review Transformers model-integration conventions. The issue names no Transformers entry point or tests, so the integration scope and validation still need to be established; done means R3M is supported using the available implementation and weights.

Written by the indexing model from the issue text.