Source: newsdio.com
In 1997, Hiroaki Kitano, a research scientist at Sony, helped organize the first Robocup, a robot soccer tournament that attracted teams of robotics and artificial intelligence researchers to compete in the picturesque city of Nagoya, Japan.
At the beginning of the first day, two teams of robots took the field. While the machines moved and examined their surroundings, a journalist asked Kitano when the game would begin. “I told him it started five minutes ago!” He says with a smile.
Such was the state of AI and robotics at that time. It took a few minutes for the machine to interpret its situation and determine what to do next. But much has changed, with artificial intelligence increasingly helping machines, from autonomous cars to surveillance cameras, to perceive and behave intelligently.
Kitano now leads a new effort at Sony, announced in November, to infuse cutting-edge artificial intelligence throughout the company. The Japanese giant believes that AI will create smarter cameras, more cunning video game characters and even the first useful kitchen robots. Kitano says that Sony believes that AI is making such rapid progress that the company needed to make technology central to its strategy.
“We have decent artificial intelligence researchers and engineers at Sony, and we have a good idea of what is happening,” says Kitano, who was attending the Conference of the Association for the Advancement of Artificial Intelligence in New York this week. “We decided that now is a moment that we should really push.”
Sony’s movement stands out among the efforts of large companies to adopt AI. It lags behind the giants of Silicon Valley in research and harnessing AI. It also has different objectives: Sony is more focused on content creation and entertainment than Google, Facebook or Apple. The Japanese giant is now looking to match the US AI titans by betting heavily on a powerful but still relatively experimental approach to AI known as reinforcement learning. Alphabet, Google’s parent company, and Amazon have also made notable investments in this technology.
Alphabet’s DeepMind used reinforcement learning to create a program capable of beating one of the best Go players in the world in 2016. Inspired by animal behavior, it involves an algorithm that refines its behavior in response to positive or negative comments.
“We believe that reinforcement learning is the same or possibly even more important” than the technologies that have driven the greatest progress in AI to date, says Kitano. “It will be the key.”
In addition to research demonstrations, reinforcement learning is being tested in areas ranging from finance to logistics. It is also emerging as a powerful way for robots to learn to deal with the real world, and to train software agents to behave intelligently in simulated environments. Therefore, it can have great potential to generate compelling video game characters and scenarios.
Reinforcement learning has been part of AI for decades, but its promise has become apparent thanks to the powerful algorithms of neural networks, modeled approximately in the way learning occurs in the brain; much more powerful computers and large amounts of training data. Even so, it is experimental and remarkably difficult to hit. Research has shown, for example, how reinforcement learning algorithms can sometimes be fixed in a reward that results in repetitive and useless behavior.
Sony will focus its artificial intelligence on three domains, says Kitano: games, sensors and, more interestingly, culinary arts. These areas reflect the company’s current commercial focus and an aspirational direction for the future.
Sony is well known for manufacturing PlayStation and games, but it also earns a large portion of its revenue from digital sensors and imaging technology. It is not difficult to see how AI could improve these areas, making games more attractive or animated or helping cameras perceive the world more intelligently.
