6 most important advances in artificial intelligence in 2016

2016 has become the year of artificial intelligence. Advances in self-government, voice recognition, and deep learning have enabled computers to make some major breakthroughs that were previously impossible. Here are six of the most significant ones.

1. AlphaGo defeated World Go Champion

Go is considered the most difficult professional game invented by mankind. It offers an incredible number of possible actions, and relies heavily on human intuition, according to the players themselves. Artificial intelligence AlphaGo trained itself to play by playing millions of games with its copies, and in March of this year managed to defeat the champion Li Sedol in four games out of five.

2. Self-driving Tesla car takes man with heart attack to hospital

Unmanned aerial vehicles are the subject of heated debate around the world. However, the head of Tesla Elon Musk emphasizes that cars with an autopilot system are safer than cars without it, regardless of the public's point of view. A report from the US National Safety Council indicates that in 2015, there were 1.3 deaths on the roads for every 100 million miles driven by conventional vehicles, while Tesla's figures were 130 million miles and just one reported accident. At the same time, on the account of the company's autopilot at least one saved life - the car delivered its owner Joshua Nelly to the hospital when he had a sudden heart attack on the way.

3. Swarm Intelligence predicted the results of the Kentucky Derby

In May, the artificial intelligence of UNU was able to successfully predict the four winners of the prestigious equestrian competition, moreover, strictly in the order of their finishing. None of the official experts in these competitions managed to do this - thus, the computer won with a bet of 540 to 1. UNU was developed by Unanimous A.I. led by human-machine interaction specialist Louis Rosenberg.

4. Microsoft AI today understands human speech better than people themselves

In October of this year, Microsoft showed that AI for the first time caught up with humans in the effectiveness of automatic speech recognition. To achieve this result, the company's system used so-called ultra-precise and recurrent neural networks. It took 2000 hours of recorded data to prepare for the test.

5. AI predicted the results of the US presidential election

The election results in America came as a surprise to many, including insiders in the political system. However, the Indian startup MogIA in Mumbai confidently predicted Trump's victory. The company's AI analyzed 20 million social media entries and was able to determine the real sympathy of voters. And although many experts are wary of attaching too much importance to the victory of MogIA, objectively - this intelligence was able to accurately predict an event that came as a surprise to billions of people.

6. Artificial intelligence has revolutionized cancer diagnostics

Healthcare is one of the areas where AI advances have the greatest practical impact. In particular, the IBM Watson supercomputer is already capable of detecting deviations in human health that elude the attention of experienced diagnosticians. Statistically, in about 30% of cases, Watson gives patients an additional diagnosis missed by human doctors.

Even more impressive results were achieved by the AI ​​of the Houston Methodist Research Institute in Texas. Artificial intelligence examines millions of mammograms (its analysis speed is 30 times faster than a human) and gives an oncological conclusion with an accuracy of 99%.