DeepMind AI learns physics by watching videos that don't make sense

South Africa News News

DeepMind AI learns physics by watching videos that don't make sense
South Africa Latest News,South Africa Headlines
  • 📰 newscientist
  • ⏱ Reading Time:
  • 76 sec. here
  • 3 min. at publisher
  • 📊 Quality Score:
  • News: 34%
  • Publisher: 51%

Teaching an AI simple physical concepts, such as the fact that two objects can't occupy the same space, could help develop more efficient algorithms

. But these models are highly specialised and lack a general understanding of the world. As DeepMind’s researchers say in their latest paper, “something fundamental is still missing”.at DeepMind and his colleagues have created an AI called Physics Learning through Auto-encoding and Tracking Objects that is designed to understand that the physical world is composed of objects that follow basic physical laws.

To test PLATO’s ability to understand five physical concepts such as persistence , solidity and unchangingness , the researchers used another series of simulated videos. Some showed objects obeying the laws of physics, while others depicted nonsensical actions, such as a ball rolling behind a pillar, not emerging from the other side, but then reappearing from behind another pillar further along its route.

They tasked PLATO to predict what would happen next in each video, and found that its predictions were reliably wrong for nonsensical videos, but usually correct for logical ones, suggesting the AI has an intuitive knowledge of physics.Piloto says the results show that an object-centric view of the world could give an AI a more generalised and adaptable set of abilities. “If you consider, for instance, all the different scenes that an apple might be in,” he says.

“That means they’re using an architecture that other people probably can’t use,” he says. “In science, it’s good to be reproducible so that other people can get the same results and then take them further.”at New York University says the findings could help to lower the computational requirements for training and running AI models.

“This is somewhat like teaching a kid what a car is by first teaching them what wheels and seats are,” he says. “The benefit of using object-centric representation, instead of raw visual inputs, makes AI learn intuitive physical concepts with better data efficiency.”

We have summarized this news so that you can read it quickly. If you are interested in the news, you can read the full text here. Read more:

newscientist /  🏆 541. in US

South Africa Latest News, South Africa Headlines

Similar News:You can also read news stories similar to this one that we have collected from other news sources.

DeepMind AI learns simple physics like a babyDeepMind AI learns simple physics like a babyNeural network could be a step towards programs for studying how human infants learn.
Read more »

Wikipedia uses Meta open-source AI to fact-check new articlesWikipedia uses Meta open-source AI to fact-check new articlesWikipedia has teamed up with Meta, the parent company of Facebook, to fact-check its citations.
Read more »

AI Learns What an Infant Knows about the Physical WorldAI Learns What an Infant Knows about the Physical WorldA computer model simulating how objects react to physical forces approximates how babies understand their surroundings
Read more »

Meta made a fact-checking AI to help verify Wikipedia citations | EngadgetMeta made a fact-checking AI to help verify Wikipedia citations | EngadgetThe Wikimedia Foundation recently partnered with Facebook parent company Meta to improve the encyclopedia’s citations..
Read more »

Scientists Have Created an AI That Can Think Like a Human BabyScientists Have Created an AI That Can Think Like a Human BabyArtificial intelligence (AI) systems are already way ahead of us in certain areas – playing Go, for example, or crunching huge sets of data – but in other aspects AI is still a long way behind human beings, even just a few months after we're born
Read more »

Intros.ai
Read more »



Render Time: 2025-03-05 01:29:28