Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
new Obstetric artificial intelligence model The company said that Meta was released this week could change how machines understand the material world, and open opportunities for the most intelligent and more robots.
The new open source style is designed, the video is called predictive structure 2, or V-jepa 2, to help artificial intelligence Mita said that understanding things like gravity and the perpetuation of the object.
“By sharing this work, we aim to give researchers and developers the best models and standards to help accelerate research and progress.” Blog post“Ultimately, which leads to the best and most capable artificial intelligence systems that will help enhance people’s lives.”
The current models that allow AI to interact with the material world depend on the data or video called the tradition of reality, but this approach emphasizes the logic of the material world, including how organisms move and interact. The model of artificial intelligence can allow concepts such as the fact that the ball rolls from a table will decrease.
Meta said the model can be useful for devices such as independent compounds and robots by ensuring that they do not need training in every possible situation. The company described it as a step towards artificial intelligence that can adapt like humans.
One of the struggle for material artificial intelligence space was the need for large amounts of training data, which takes time, money and resources. In SXSW earlier this year, experts said the artificial data – training data created by artificial intelligence – could help Prepare a more traditional educational model For unexpected cases. (In Austin, the example used was the appearance of the bats from the famous congressional bridge in the city.)
Meta said that her new model simplifies the process and makes it more efficient in realistic applications because it does not depend on all these training data.
The following steps for the world models include training models capable of learning, thinking and planning through different standards of time and space, which makes them better in breaking the complex tasks. Multimedia models, which can use other senses such as sound and touch in addition to vision, will also help artificial intelligence models in the future to understand the real world.
Watch this: These AI robots want to do your business for you