A DeepMind presence in the US will aid collaboration with parent company Google
Alphabet’s London-based artificial intelligence (AI) division DeepMind has signalled its first forays into the US after advertising for a researcher to be based at Google’s Mountain View headquarters in California.
The “applied research scientist position” will apparently be the first in “a couple dozen” new hires as DeepMind looks to build a small team in the States to “bridge the gap between Google and our team in London, helping us collaborate even more closely to bring our research breakthroughs to Google users around the world.”
The job listing continues: “The Applied team in Mountain View will be made up of a mixture of software engineers and research scientists who work together to solve real-world problems at Google-scale.”
The new personnel will be tasked with making reinforcement learning , i.e. using rewards to teach good behaviour, more applicable to solving real-world problems and will have to compete with researchers from Google Brain, Google’s other AI division which also has offices at the Mountain View campus.
As well as increasing its scale, a US presence suggests that Google might be looking to add more smart software into its services in conjunction with the increased prevalence of AI in our society.
DeepMind – purchased by Google for $400 million (£242m) in 2014 – currently employs around 400 people at its King’s Cross office. They are split into two divisions; one which focuses on computer science research and an ‘applied division’ which builds AI-based products and services designed for real-world applications.
The DeepMind team was the centre of some controversy earlier in the year after signing an agreement with the NHS, giving it access to the sensitive data of around 1.6 million patients from three NHS Trust hospitals.
Since then, it has worked with Google to slash the amount of power needed to cool its data centres by 40 percent, created a system that uses deep learning to navigate the London underground and created a computer model that generates natural sounding speech and music.