Google DeepMind’s AlphaGo victory not ‘true AI’, says Facebook’s AI chief
Facebook’s director of AI research says it’s ‘completely, utterly, ridiculously wrong’ to think AlphaGo beating Lee Sedol was ‘true artificial intelligence’
Short of time?
Facebook's director of AI research, Yann LeCun, has dismissed Google DeepMind’s victory over Go world champion Lee Sedol as ‘not true artificial intelligence’.
Google DeepMind made history last week when its AI program AlphaGo became the first computer in history to defeat a top-ranked human player of the ancient Chinese board game.
Sedol scored his first win over the computer yesterday, but with AlphaGo going 3-0 up last week, the five-match series was already wrapped up. Google has since enjoyed resounding plaudits from the global technology community for its advancements in AI.
However, LeCun has taken exception to a post about AlphaGo’s victory on technology news website Slashdot, which said: 'We know now that we don't need any big new breakthroughs to get to true AI.’
‘That is completely, utterly, ridiculously wrong,’ LeCun blasted. ‘As I've said in previous statements: most of human and animal learning is unsupervised learning. If intelligence was a cake, unsupervised learning would be the cake, supervised learning would be the icing on the cake, and reinforcement learning would be the cherry on the cake.
‘We know how to make the icing and the cherry, but we don't know how to make the cake. We need to solve the unsupervised learning problem before we can even think of getting to true AI. And that's just an obstacle we know about. What about all the ones we don't know about?’
LeCun also linked to an article by John Langford, a machine learning researcher at Microsoft, claiming 'AlphaGo is not the solution to AI'. 'Words of wisdom,' LeCun commented.
LeCun heads up the Facebook Artificial Intelligence Research (FAIR) group, which leads the social media giant’s efforts in advancing the field of machine intelligence and building 'the best AI lab in the world'.
But despite posting developments in January of its own progress in using AI to play Go, Facebook’s AI has yet to beat a professional player.
Facebook’s founder and CEO Mark Zuckerberg appeared to take a different view to his AI chief, congratulating Google DeepMind on a ‘historic milestone’ in AI research.
Facebook and Google have both invested heavily in AI research, and Zuckerberg's comments could be perceived as an open challenge for his own team to up their game.
DeepMind Technologies was founded in 2010 by Mustafa Suleyman and Demis Hassabis from the UK, and New Zealander Shane Legg, who Hassabis met at University College London's Gatsby Computational Neuroscience Unit.
Using AI, the start-up built a computer that mimics the short-term memory of the human brain.
In 2014, Google acquired the company for £400 million and renamed it Google DeepMind, but it has remained headquartered in London.
Facebook’s own efforts emanate back to December 2013, when LeCun revealed Zuckerberg had poached him to head up a new research lab with the 'ambitious, long-term goal of bringing about major advances in AI'.
IBM made headlines when its AI computer Deep Blue defeated chest grandmaster Garry Kasparov in 1997, but chess only has around 10 to the power of 60 possible ways a game can be played, compared to 10 to the power of 700 possible scenarios in Go.