Skip to main content

Facebook taps Minecraft as training ground for next stage of A.I.

Facebook researchers have chosen Minecraft as the training ground for the development of the next stage of artificial intelligence, as the technology looks to conquer a major challenge.

A.I. systems have been learning to carry out specific tasks, including playing soccer and filling in gaps in images, and have proven to be better at some of them than humans, such as in games such as in Texas Hold ’em poker and Quake III‘s Capture the Flag mode.

One of the biggest limitations of the current forms of A.I. is that, while they may excel at a specific activity, they are unable to deal with multiple tasks. Facebook Research’s Arthur Szlam and his colleagues, who have started working on an A.I. assistant that can perform a variety of tasks, have decided to break through the limitation with the help of Minecraft.

Minecraft, possibly the best-selling video game of all time, allows players to move around a 3D environment, exploring and building in a limitless world. The game’s infinite variety, in combination with simple and predictable rules, makes it an ideal environment for training A.I.

The goal of Szlam and his team is to create an A.I. assistant that is capable of helping people with different kinds of tasks, and they believe that Minecraft will help them achieve that. A generalist A.I. assistant will be more useful to the regular user, compared to an A.I. system that can only do one thing well.

“Instead of superhuman performance on a single difficult task, we are interested in competency across a large number of simpler tasks, specified (perhaps poorly) by humans,” Szlam and his team wrote in their research.

Within Minecraft, the A.I. assistant will need to learn the various concepts of the game and the nature of the possible requests. The requests may become very complex, but the opportunity for learning is massive, with the possibility of pushing A.I. research to the next level.

Szlam and his team have uploaded data and code for a baseline Minecraft assistant that is available to try. Perhaps this is a very early look at the generalist A.I. that will be the future of the technology.

Editors' Recommendations

Aaron Mamiit
Aaron received a NES and a copy of Super Mario Bros. for Christmas when he was 4 years old, and he has been fascinated with…
How Nvidia is using A.I. to help Domino’s deliver pizzas faster
Domino's delivery in line.

Nvidia announced a new tool that can help deliver your pizzas faster -- yes, really -- at its fall GTC 2021 event. It's called ReOpt, and it's a real-time logistics tool that Domino's is already using to optimize delivery routes based on time and cost.

ReOpt is a set of logistics-planning algorithms that can find billions of routes to the same location. It utilizes heuristics powered by GPU computing to route vehicles in the most efficient way possible. It's like Google Maps, just way more complex and designed specifically to meet the needs of last-mile delivery.

Read more
The funny formula: Why machine-generated humor is the holy grail of A.I.
microphone in a bar

In "The Outrageous Okona," the fourth episode of the second season of Star Trek: The Next Generation, the Enterprise's resident android Data attempts to learn the one skill it has previously been unable to master: Humor. Visiting the ship’s Holodeck, Data takes lessons from a holographic comedian to try and understand the business of making funny.

While the worlds of Star Trek and the real world can be far apart at times, this plotline rings true for machine intelligence here on Earth. Put simply, getting an A.I. to understand humor and then to generate its own jokes turns out to be extraordinarily tough.

Read more
Nvidia’s latest A.I. results prove that ARM is ready for the data center
Jensen Huang at GTX 2020.

Nvidia just published its latest MLPerf benchmark results, and they have are some big implications for the future of computing. In addition to maintaining a lead over other A.I. hardware -- which Nvidia has claimed for the last three batches of results -- the company showcased the power of ARM-based systems in the data center, with results nearly matching traditional x86 systems.

In the six tests MLPerf includes, ARM-based systems came within a few percentage points of x86 systems, with both using Nvidia A100 A.I. graphics cards. In one of the tests, the ARM-based system actually beat the x86 one, showcasing the advancements made in deploying different instruction sets in A.I. applications.

Read more