Notebookcheck Logo
, , , , , ,
search relation.
, , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,

University of California-developed AI taught itself to solve a Rubik's Cube in one second

Image via Nature Machine Intelligence
Image via Nature Machine Intelligence
Researchers at the University of California created an AI that taught itself the shortest solution paths in several different one-goal logic puzzles, including a Rubik's Cube, sliding tile puzzles, Lights out, and Sokoban. While the AI is much slower than human-programmed robots, it gives an interesting look into how AIs arrive at solutions to problems.
Sam Medley,

Researchers at the University of California developed an artificial intelligence that taught itself to solve a 3x3x3 Rubik’s Cube in just over a second. While not the fastest machine to solve a Rubik’s Cube (that title is currently held by a robot that solved one in 0.8 seconds), it is an interesting look into how machine learning tackles logic problems and puzzles.

The AI, dubbed DeepCubeA, ran through 10 billion different start states of the Rubik’s Cube and was instructed to solve the Cube in 30 moves or fewer. The AI did relatively well, solving 1000 tested iterations. Perhaps more impressively (from a machine learning view), DeepCubeA found the shortest solution path 60.3% of the time. Keep in mind that this was achieved solely by the algorithm running within DeepCubeA; outside of creating the original AI, there was no outside assistance given to the machine.

The research team also tested DeepCubeA on other algorithmically-based logic puzzles with one goal state. Notably, the team ran DeepCubeA through multiple iterations of sliding tile puzzles with 15 tiles, 24 tiles, 35 tiles, and 48 tiles; a 7x7 game of Lights Out; and the box-pushing game Sokoban. DeepCubeA performed well in most tests, beating other algorithmic systems in each puzzle.

The full article detailing the research methodology was published in a recent issue of Nature Machine Intelligence. It is an interesting look into how far AI has come. Perhaps we will soon see an artificial intelligence that can find optimal solutions to simple real-world problems, laying the groundwork for complex learning algorithms that could control everyday systems. We’ll leave it to you to decide whether that is a dream or dystopian future

static version load dynamic
Loading Comments
Comment on this article
Sam Medley
Sam Medley - Senior Tech Writer - 1171 articles published on Notebookcheck since 2016
I've been a computer geek my entire life. After graduating college with a degree in Mathematics, I worked in finance and banking a few years before taking a job as a database administrator. I started working with Notebookcheck in October of 2016 and have enjoyed writing news and reviews. I've also written for other outlets including UltrabookReview and GeeksWorldWide, focusing on consumer guidance and video gaming. My areas of interest include the business side of technology, retro gaming, Linux, and innovative gadgets. When I'm not writing on electronics or tinkering with a device, I'm either outside with my family, enjoying a decade-old video game, or playing drums or piano.
Please share our article, every link counts!
> Notebook / Laptop Reviews and News > News > News Archive > Newsarchive 2019 07 > University of California-developed AI taught itself to solve a Rubik's Cube in one second
Sam Medley, 2019-07-16 (Update: 2019-07-16)