So what happens when you teach artificial intelligence (AI) to be curious?
It seems that it becomes a telly-addict, according to the latest AI researchers exploring how to give algorithms a sense of curiosity, so they can learn without any human guidance.
New research from OpenAI, the non-profit Artificial Intelligence lab founded by Elon Musk, Sam Altman, and other Silicon Valley players, in collaboration with researchers from UC Berkeley and the University of Edinburgh, has found that when an AI algorithm is given a simple definition of curiosity, it can, without any human-provided information, explore over 50 video games and even beat some of them.
Yet it also loves watching dead TV channels and mindlessly flicking between channels, which is something we are all guilty of, right?
The researchers found that because the AI agent was rewarded for seeing new things, sometimes it would die on purpose just to see the Game Over screen on a game, or just tune in to fake TV and its remote, flipping through channel after channel to discover something to satiate its curiosity.
The OpenAI team’s artificial curiosity algorithm would try to predict what its environment would look like one frame into the future.
OpenAI researcher Harri Edwards says that getting the AI agent to flip through channels came from a thought experiment called the noisy-TV problem. The static on a TV is immensely random, so a curious AI agent could never truly predict what would happen next, and get drawn into watching the TV forever.
The researchers tested their theory by putting a digital TV inside a 3D environment, and allowing the agent to press a button to change the channel. When the agent found the TV and started flipping through the channels, the stream of new images made the TV irresistible.
Edwards said that the AI could pry itself away from the TV only when the AI’s surroundings somehow seemed more interesting than the next thing on TV.
Seems AI is as bored as humans more than we expect!