[ad_1]
Many of us remember the famous scene in Iron Man 2 movie where Tony Stark’s AI system JARVIS helped him create a new element. Google has achieved something similar in real life. Google DeepMind, the company’s AI arm has a new use case for artificial intelligence – predicting the structure of materials that are not yet discovered or known.
Google DeepMind has created a system, known as the A-Lab, that can predict the structure of materials, which later can be created in labs for use in improving real-world technologies such as batteries or solar cells.
The system can carry out the synthesis and analysis of the products — all without human intervention, the company said in a research paper published in journal Nature. The Alphabet-owned AI firm said the tool has predicted nearly 400,000 stable substances which could be produced in lab conditions.
Meanwhile, another AI system has predicted the existence of hundreds of thousands of stable materials.
“A lot of the technologies around us, including batteries and solar cells, could really improve with better materials,” said Ekin Dogus Cubuk, who leads the materials discovery team at Google DeepMind in London.
Why it is important discovery
The development holds much importance as it could help engineers help in discovery and synthesis of new materials, which can be a costly and time-consuming process, in much less time.
“We’re hoping that big improvements in experimentation, autonomous synthesis, and machine learning models will significantly shorten that 10 to 20-year timeline to something that’s much more manageable,” Cubuk added.
DeepMind’s AI was trained on data from the Materials Project, an international research group founded at the Lawrence Berkeley National Laboratory in 2011. The project is made up of existing research of around 50,000 already-known materials.
The company said it will share its data with the research community to further accelerate breakthroughs in material discovery.
Google DeepMind has created a system, known as the A-Lab, that can predict the structure of materials, which later can be created in labs for use in improving real-world technologies such as batteries or solar cells.
The system can carry out the synthesis and analysis of the products — all without human intervention, the company said in a research paper published in journal Nature. The Alphabet-owned AI firm said the tool has predicted nearly 400,000 stable substances which could be produced in lab conditions.
Meanwhile, another AI system has predicted the existence of hundreds of thousands of stable materials.
“A lot of the technologies around us, including batteries and solar cells, could really improve with better materials,” said Ekin Dogus Cubuk, who leads the materials discovery team at Google DeepMind in London.
Why it is important discovery
The development holds much importance as it could help engineers help in discovery and synthesis of new materials, which can be a costly and time-consuming process, in much less time.
“We’re hoping that big improvements in experimentation, autonomous synthesis, and machine learning models will significantly shorten that 10 to 20-year timeline to something that’s much more manageable,” Cubuk added.
DeepMind’s AI was trained on data from the Materials Project, an international research group founded at the Lawrence Berkeley National Laboratory in 2011. The project is made up of existing research of around 50,000 already-known materials.
The company said it will share its data with the research community to further accelerate breakthroughs in material discovery.
[ad_2]
Source link
