>
New covid-19 guidance from CDC focuses on individual decisions
Corona-Warn-App: Vaccination status will be recognizable by color
Interest-free loans to be rolled out in UK to help with food bills
China to send troops to Russia for 'Vostok' exercise
What's EASIER? MILKING a MINI COW or FULL SIZED COW?
Fully Charged Checks Out Aptera, Drives The Three-Wheeler Solar EV
This Man Built His Own ISP. Now He's Getting $2.6M to Expand It
Blowhole wave energy generator exceeds expectations in 12-month test
3-wheeled EV commuter equals 230 MPGe, blends torque & safety
Starlink Wins FCC Approval For In-Motion Use On Airplanes And Cruise Ships
Raspberry Pi Foundation brings Wi-Fi to Pico microcontroller
Have You Changed Phones Yet?, + Q&A
Breakthrough Zero-Carbon Fertilizer Set to Take Root Across the World as 'Biochar'
Artificial Photosynthesis Can Produce More Food in the Dark Than With Sunshine
The agent, which Deepmind refer to as Gato, works as a multi-modal, multi-task, multi-embodiment generalist policy. The same network with the same weights can play Atari, caption images, chat, stack blocks with a real robot arm and much more, deciding based on its context whether to output text, joint torques, button presses, or other tokens. In this report Deepmind describe the model and the data, and document the current capabilities of Gato.
A generalist agent. Gato can sense and act with different embodiments across a wide range of environments using a single neural network with the same set of weights. Gato was trained on 604 distinct tasks with varying modalities, observations and action specifications.
Transformer sequence models are effective as multi-task multi-embodiment policies, including for real-world text, vision and robotics tasks. They show promise as well in few-shot out-of-distribution task learning. In the future, such models could be used as a default starting point via prompting or fine-tuning to learn new behaviors, rather than training from scratch.
Given scaling law trends, the performance across all tasks including dialogue will increase with scale in parameters, data and compute. Better hardware and network architectures will allow training bigger models while maintaining real-time robot control capability. By scaling up and iterating on this same basic approach, Deepmind can build a useful general-purpose agent.