In sci-fi films, robots are depicted as some kind of futuristic creatures that can walk on two legs and think like people. However, the cinematic sketches are not entirely accurate. In fact, we have been using robots for a long time, but they look a little different. These are looms of the early 19th century, cruise missiles from the Second World War, modern dishwashers. Even our alarms are small jobs that follow a set program to wake us up at the right time. However, as the history of robotics shows, we are not too far from the sci-fi reality. Here are 5 of the smartest human-made robots in the last 50 years.
Eliza, 1966
In sci-fi films, robots are depicted as some kind of futuristic creatures that can walk on two legs and think like people. However, the cinematic sketches are not entirely accurate. In fact, we have been using robots for a long time, but they look a little different. These are looms of the early 19th century, cruise missiles from the Second World War, modern dishwashers. Even our alarms are small jobs that follow a set program to wake us up at the right time. However, as the history of robotics shows, we are not too far from the sci-fi reality. Here are 5 of the smartest human-made robots in the last 50 years.
Virtual Beings, 1994
Computer graphics specialist and researcher Karl Sims created a group of virtual beings that inhabited their own virtual universe. Thanks to genetic algorithms, they learned to swim, crawl, jump and compete with each other. Unfortunately, they were too worried about their own virtual lives to talk to us.
Promotional video:
Deep Blue, 1997
IBM's Deep Blue supercomputer rose to prominence after winning a six-game match against world chess champion Garry Kasparov. No robot has ever achieved similar results. However, Deep Blue is not able to do without the help of programmers, and using only the basic methods of the theory of artificial intelligence, he can hardly fully reflect on his actions. After a crushing defeat, Garry Kasparov accused his opponent of foul play and demanded a rematch, but the developer company refused the chess player.
I BM Watson, 2011
IBM Watson became the first supercomputer to defeat a human in the American TV quiz game Jeopardy! A smart computer could understand the questions asked and find answers to them in a database of about 200 million pages with information totaling 4 terabytes. The car played against Ken Jennings, who set the record for the longest unbeaten streaks, and Brad Rutter, who won the biggest win in the intellectual show. IBM Watson prevailed and won $ 1 million, while its rivals received 300 and 200 thousand, respectively.
Deep-Q Network, 2015
Deep-Q Network was developed by DeepMind, which created the AlphaGo program, which defeated the champion of the Japanese strategy game go. The artificial intelligence learned the rules of 49 classic Atari games on its own, just by looking at the screen without any help from the programmers. Deep-Q Network handled most games brilliantly. However, he was never able to surpass the human in the Pac-Man game. Deep-Q Network was able to learn the rules, but it was unable to plan its actions for a few seconds in advance.