Skip to main content

AI learns how to tackle new situations by studying how humans play games

nestor ai paying attention artificial intelligence
Image used with permission by copyright holder
If artificial intelligence is going to excel at driving cars or performing other complex tasks that we humans take for granted, then it needs to learn how to respond to unknown circumstances. That is the task of machine learning, which needs real-world examples to study.

So far, however, most data used to train machine-learning systems comes from virtual environments. A group of researchers, including a Microsoft Research scientist from the U.K., have set out to change that by using game replay data that can show an AI how humans tackle complex problems.

The researchers used Atari 2600 game replays to provide real-world data to a deep learning system that uses trial and error, or reinforcement learning (RL), to tackle new tasks in a previously unknown environment. The data used in the study represents what the researchers called the “largest and most diverse such data set” that has ever been publicly released.

The data was gathered by making a web-based Atari 2600 emulator, called the Atari Grand Challenge, available using the Javatari tool written in Javascript. The researchers used a form of gamified crowdsourcing, which leveraged people’s desire to play games in order to be helpful along with a reward mechanism that ranked each player’s performance.

Around 9.7 million frames or about 45 hours of gameplay time were collected and analyzed. Five games were used in creating the data based on their varying levels of difficulty and complexity: Video Pinball, Qbert, Space Invaders, Ms. Pac-Man, and Montezuma’s Revenge.

The results have been promising so far. By feeding information into the system like player actions taken during the games, in-game rewards, and current scores, the researchers were able to demonstrate the value of using this kind of data to train machine learning systems. Going forward, the researchers hope to use professional players to improve the data’s ability to train AI that is even better at responding to unknown situations.

Editors' Recommendations

Mark Coppock
Mark has been a geek since MS-DOS gave way to Windows and the PalmPilot was a thing. He’s translated his love for…
This new Photoshop tool could bring AI magic to your images
A mountainous landscape at night with the Northern Lights in the sky, a lake in the foreground, and a person standing under a rock archway on the right. This image was made with Adobe Photoshop's Generative Fill tool.

These days, it seems like everyone and their dog is working artificial intelligence (AI) into their tech products, from ChatGPT in your web browser to click-and-drag image editing. The latest example is Adobe Photoshop, but this isn’t just another cookie-cutter quick fix -- no, it could have a profound effect on imagery and image creators.

Photoshop’s newest feature is called Generative Fill, and it lets you use text prompts to automatically adjust areas of an image you are working on. This might let you add new features, adjust existing elements, or remove unwanted sections of the picture by typing your request into the app.

Read more
OpenAI’s new ChatGPT app is free for iPhone and iPad
The ChatGPT website on an iPhone.

OpenAI has just launched a free ChatGPT app for iOS, giving iPhone and iPad owners an easy way to take the AI-powered tool for a spin.

The new app, which is able to converse in a remarkably human-like way, is available now in the U.S. App Store and will come to additional countries “in the coming weeks,” OpenAI said. Android users are promised their own ChatGPT app “soon.”

Read more
New AI tools could ‘easily’ lead to 4-day week, expert says
Dell's new UltraSharp monitors are modular video conferencing solutions.

ChatGPT and similar generative artificial intelligence (AI) tools are only going to get better, with many experts envisaging a major shake-up for white-collar professions in the coming years.

The new wave of AI-powered chatbots that have been garnering so much attention in recent months can converse in an impressively human-like way and before long will be able to comfortably handle numerous tasks across a range of industries.

Read more