One step closer to the Matrix: AI defeats human champion in Street Fighter — with a revolutionary type of memory it uses to make itself even more powerful

gameplay
(Image credit: Capcom)

Researchers from the Singapore University of Technology and Design (SUTD) created a new software centered around reinforcement learning and phase-change memory that’s designed to understand complicated movement design.

Previous work has applied this kind of deep learning to other games like Chess or Go, but they decided instead to expose the D-PPO algorithm to the rigors of Street Fighter Champion Edition II. The SUTD researchers trained its SF-R2 AI player on two days of consecutive play against the computer, before letting it loose on a human participant – who the AI-powered system beat comfortably. 

The work has implications for movement science more broadly, according to the research paper, and can possibly be fed into improving robotics and autonomous vehicles, for example. It paves the way for broadly applicable training in fields where machines may observe human norms and attempt to replicate and outperform them.

Ready Pl-AI-yer One

One of the major milestones that AI researchers have used to measure the effectiveness of the systems they’ve built is by letting them compete with human players in different kinds of games. This has been happening for some time.

In 2017, an Alpha Go AI built by DeepMind beat the number-one human Go player in the world for the second time, following the first victory over Fan Hui the previous year. Microsoft’s AI, in June, achieved the world’s first perfect Ms. Pac-Man score, and in August we saw an OpenAI engine beating the best Dota 2 players of the time.

This latest milestone – besting a Street Fighter champion – was made possible due to reinforcement learning as well as phase-change memory. First developed by HP, this is a form of nonvolatile memory achieved by using electrical charges to change areas on chalcogenide glass. It’s much faster than commonly used Flash memory.

"Our approach is unique because we use reinforcement learning to solve the problem of creating movements that outperform those of top human players,” said principal investigator Desmond Loke to TechXplore. “This was simply not possible using prior approaches, and it has the potential to transform the types of moves we can create.

More from TechRadar Pro

Keumars Afifi-Sabet
Channel Editor (Technology), Live Science

Keumars Afifi-Sabet is the Technology Editor for Live Science. He has written for a variety of publications including ITPro, The Week Digital and ComputerActive. He has worked as a technology journalist for more than five years, having previously held the role of features editor with ITPro. In his previous role, he oversaw the commissioning and publishing of long form in areas including AI, cyber security, cloud computing and digital transformation.

Read more
Two business men playing chess in the office.
It turns out ChatGPT o1 and DeepSeek-R1 cheat at chess if they’re losing, which makes me wonder if I should trust AI with anything
A hand reaching out to touch a futuristic rendering of an AI processor.
DeepSeek and the race to surpass human intelligence
Humanity's Last Exam
OpenAI's Deep Research smashes records for the world's hardest AI exam, with ChatGPT o3-mini and DeepSeek left in its wake
A person's hand using DeepSeek on their mobile phone
'A virtual DPU within a GPU': Could clever hardware hack be behind DeepSeek's groundbreaking AI efficiency?
Nvidia HQ
Nvidia calls DeepSeek an 'excellent AI advancement' and praises the Chinese AI app's ingenuity
Nvidia H800 GPU
A look at the unbelievable Nvidia GPU that powers DeepSeek's AI global ambition
Latest in Pro
An image of network security icons for a network encircling a digital blue earth.
Why multi-CDNs are going to shake up 2025
URL phishing
HaveIBeenPwned owner suffers phishing attack that stole his Mailchimp mailing list
Ransomware
Cl0p resurgence drives ransomware attacks to new highs in 2025
Millwall FC The Den
The UK's first football club mobile network is here - but you probably won't guess which team has launched it
A person using a smartphone with a cybersecurity lock symbol appearing over it.
The growing threat of device code phishing and how to defend against It
Cybersecurity
Why OT security needs exposure management to break the cycle of endless patching
Latest in News
Microsoft Surface Laptop and Surface Pro devices on a table.
Hate Windows 11’s search? Microsoft is fixing it with AI, and that almost makes me want to buy a Copilot+ PC
Oura Ring 4
Activity tracking on Oura Ring is about to get a whole lot better, but I've got bad news about your step count
Google Pixel Buds Pro 2
Cleaned your Pixel Buds Pro 2 recently? If not, you might be getting worse sound
Google Maps on a phone being held in someone's hand
Google Maps is getting two key upgrades, for easier route planning and quicker access to Gemini AI
URL phishing
HaveIBeenPwned owner suffers phishing attack that stole his Mailchimp mailing list
Gemini on a smartphone.
Gemini 2.5 is now available for Advanced users and it seriously improves Google’s AI reasoning