Robotics at Google
PaLM-SayCan Demo
Language as Robot Middleware
Andy Zeng, Fei Xia
CoRL Workshop on Language and Robot Learning
Manipulation
TossingBot
Interact with the physical world to learn bottom-up commonsense
Transporter Nets
Implicit Behavior Cloning
w/ machine learning
i.e. "how the world works"
On the quest for shared priors
Interact with the physical world to learn bottom-up commonsense
w/ machine learning
i.e. "how the world works"
# Tasks
Data
On the quest for shared priors
Interact with the physical world to learn bottom-up commonsense
w/ machine learning
i.e. "how the world works"
# Tasks
Data
Expectation
Reality
Complexity in environment, embodiment, contact, etc.
Deep Learning is a Box
Interpolation
Extrapolation
adapted from Tomás Lozano-Pérez
Deep Learning is a Box
Interpolation
Extrapolation
Roboticist
Vision
NLP
Deep Learning is a Box
Interpolation
Extrapolation
Internet
Meanwhile in NLP...
Large Language Models
Large Language Models?
Internet
Meanwhile in NLP...
Books
Recipes
Code
News
Articles
Dialogue
Demo
Quick Primer on Language Models
Tokens (inputs & outputs)
Transformers (models)
Attention Is All You Need, NeurIPS 2017
Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, Illia Polosukhin
Quick Primer on Language Models
Tokens (inputs & outputs)
Transformers (models)
Pieces of words (BPE encoding)
big
bigger
per word:
biggest
small
smaller
smallest
big
er
per token:
est
small
Attention Is All You Need, NeurIPS 2017
Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, Illia Polosukhin
Quick Primer on Language Models
Tokens (inputs & outputs)
Transformers (models)
Self-Attention
Pieces of words (BPE encoding)
big
bigger
per word:
biggest
small
smaller
smallest
big
er
per token:
est
small
Attention Is All You Need, NeurIPS 2017
Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, Illia Polosukhin
Bigger is Better
Neural Language Models: Bigger is Better, WeCNLP 2018
Noam Shazeer
Bigger is Better
Neural Language Models: Bigger is Better, WeCNLP 2018
Noam Shazeer
Bigger is Better
Neural Language Models: Bigger is Better, WeCNLP 2018
Noam Shazeer
Bigger is Better
Neural Language Models: Bigger is Better, WeCNLP 2018
Noam Shazeer
Bigger is Better
Neural Language Models: Bigger is Better, WeCNLP 2018
Noam Shazeer
Robot Planning
Visual Commonsense
Robot Programming
Socratic Models
Code as Policies
PaLM-SayCan
Demo
Somewhere in the space of interpolation
Lives
Somewhere in the space of interpolation
Robot Planning
Visual Commonsense
Robot Programming
Socratic Models
Code as Policies
PaLM-SayCan
Demo
Lives
PaLM-SayCan
Live Demo
Fei Xia
LLM planning on robots! Open research problem, but here's one way to do it
Do As I Can, Not As I Say: Grounding Language in Robotic Affordances
https://say-can.github.io
PaLM
Code is available
Limits of language as information bottleneck?
- Loses spatial precision
- Highly (distributional) multimodal
- Not as information-rich as e.g. images
- Only for high level? what about control?
Perception
Planning
Control
Socratic Models
Inner Monologue
SayCan
Wenlong Huang et al, 2022
Imitation? RL?
Engineered?
Limits of language as information bottleneck?
- Loses spatial precision
- Highly (distributional) multimodal
- Not as information-rich as e.g. images
- Only for high level? what about control?
Perception
Planning
Control
Socratic Models
Inner Monologue
SayCan
Wenlong Huang et al, 2022
Imitation? RL?
Engineered?
Scaling robot learning (today) is hard
Robot Learning
Not a lot of robot data
500 expert demos
5000 expert demos
50 expert demos
Robot Learning
Language Models
Not a lot of robot data
Lots of Internet data
500 expert demos
5000 expert demos
50 expert demos
256K token vocab w/ word embedding dim = 18,432
PaLM-sized robot dataset = 100 robots for 24 yrs
collecting (mostly) diverse data
Scaling robot learning (today) is hard
Robot Learning
Language Models
- Finding other sources of data (sim, YouTube)
- Improve data efficiency with prior knowledge
Not a lot of robot data
Lots of Internet data
500 expert demos
5000 expert demos
50 expert demos
256K token vocab w/ word embedding dim = 18,432
PaLM-sized robot dataset = 100 robots for 24 yrs
collecting (mostly) diverse data
Scaling robot learning (today) is hard
Robot Learning
Language Models
- Finding other sources of data (sim, YouTube)
- Improve data efficiency with prior knowledge
Not a lot of robot data
Lots of Internet data
500 expert demos
5000 expert demos
50 expert demos
Use language to help close the gap!
256K token vocab w/ word embedding dim = 18,432
PaLM-sized robot dataset = 100 robots for 24 yrs
collecting (mostly) diverse data
Scaling robot learning (today) is hard
2 Final Thoughts
adapted from Tomás Lozano-Pérez
Machine learning is a box
... but robotics is a line
Embrace compositionality
adapted from Tomás Lozano-Pérez
Machine learning is a box
... but robotics is a line
Embrace compositionality
adapted from Tomás Lozano-Pérez
Machine learning is a box
... but robotics is a line
2. compose them
autonomously
1. build boxes
"Language" as the glue for intelligent machines
Language
Perception
Planning
Control
Socratic Models: Composing Zero-Shot Multimodal Reasoning with Language
https://socraticmodels.github.io
We have some reason to believe that
"the structure of language is the structure of generalization"
To understand language is to understand generalization
https://evjang.com/2021/12/17/lang-generalization.html
Sapir–Whorf hypothesis
Towards grounding everything in language
Language
Control
Vision
Tactile
Socratic Models: Composing Zero-Shot Multimodal Reasoning with Language
https://socraticmodels.github.io
Lots of data
Less data
Less data
Thank you!
Pete Florence
Adrian Wong
Johnny Lee
Vikas Sindhwani
Stefan Welker
Vincent Vanhoucke
Kevin Zakka
Michael Ryoo
Maria Attarian
Brian Ichter
Krzysztof Choromanski
Federico Tombari
Jacky Liang
Aveek Purohit
Wenlong Huang
Fei Xia
Peng Xu
Karol Hausman
and many others!