AI Algorithm Can Now Solve Rubik’s Cube In Less Than a Second
Researchers have created an AI algorithm that can solve a Rubiks Cube in a fraction of a second, quicker than most people. The work is a step towards creating AI systems that can think, reason, plan, and make decisions. The study, released in the journal Nature Machine Intelligence, demonstrates that DeepCubeA, a profound reinforcement learning algorithm programmed by computer scientists and mathematicians at the University of California, can solve the Rubik’s Cube in a fraction of a second without any particular domain knowledge or human in-game training.
This is not a simple task considering that the cube has completion paths numbering in the billions but only one goal state-each of six sides showing a solid color-which apparently cannot be discovered through random moves.
“Artificial Intelligence can defeat the world’s best human chess and Go players, but some of the more difficult puzzles, such as the Rubik’s Cube, had not been solved by computers, so we thought they were open for AI approaches,” said study author Pierre Baldi, Professor at the University of California.
“The solution to the Rubik’s Cube involves more symbolic, mathematical and abstract thinking, so a deep learning machine that can crack such a puzzle is getting closer to becoming a system that can think, reason, plan and make decisions,” Baldi said.
The scientists proved that DeepCubeA solved 100 percent of all test configurations for the research, finding the fastest route to goal state about 60 percent of the time.
The algorithm also operates on other combinatorial games such as the puzzle of sliding tiles, Lights Out and Sokoban.
The researchers were interested in knowing how and why Artificial Intelligence (AI) created its moves and how long it took to perfect its technique.
“It learned on its own, our AI takes about 20 moves, most of the time solving it in the minimum number of steps,” Baldi said.
“Right there, you can see the strategy is different, so my best guess is that the AI’s form of reasoning is completely different from a human’s,” he added.
According to the researchers, the ultimate goal of such projects is to build the next generation of AI systems.