LISTEN NOW: SPRYTE WEATHER DIARY (episode: TUESDAY 28th APRIL - Fictitious Weather, Trampoline, Ferrari, Diabolical Chaos, Like A Taurus)

SUNDAY 10 MAY 2026
12:00 – 00:00 (BST)

Outplaying Elite Table Tennis Players with an Autonomous Robot

Today we'll be showing a running loop of videos from the supplementary material released as part of a recently published paper on current developments of Ace, an autonomous system which is designed to operate competitively with elite human table tennis players.

The video footage will play back-to-back, on loop for 12 hours from midday to midnight (BST) on our website.

This stream is of course not in collaboration or direct connection with Sony or any other institution affiliated with the research. We both just think the video footage is fascinating (both on its own and as a visual representation of this acute overlap of machine learning, robotics, high-speed sports, lab experimentation… and the compressing of all that within these videos of real-time play) and it's something we really want to share through our particular way of one-off, 12-hour streams.

Here's the abstract from the paper to give an idea of the study which led to this collection of footage that we'll be streaming on Sunday:

Outplaying elite table tennis players with an autonomous robot
Dürr, P., El Gheche, M., Maeda, G.J. et al.

Abstract

Artificial intelligence (AI) systems now challenge or surpass human experts in many computer games1,2. Physical and real-time sports such as table tennis, however, remain a major open challenge because of their requirements for fast, precise and adversarial interactions near obstacles and at the edge of human reaction time3. Here we present Ace, to our knowledge the first real-world autonomous system competitive with elite human table tennis players. Ace addresses the challenges of physical real-time interaction through a new, high-speed perception system using event-based vision sensors4, and a new control system based on model-free reinforcement learning, as well as state-of-the-art high-speed robot hardware. Evaluated in matches against elite and professional players under official competition rules, Ace achieved several victories and demonstrated consistent returns of high-speed, high-spin shots. These results highlight the potential of physical AI agents to perform complex, real-time interactive tasks, suggesting broader applications in domains requiring fast, precise human–robot interaction.

Available via Open Access

CC BY-NC-ND 4.0

Dürr, P., El Gheche, M., Maeda, G.J. et al. Outplaying elite table tennis players with an autonomous robot. Nature 652, 886–891 (2026). https://doi.org/10.1038/s41586-026-10338-5

Spryte is an online artist-led project. We show artists' moving-image works, as well as audio and text. These all take place sporadically online and are free to access. Spryte is fundamentally artist-run, we don’t really have any fixed programmes and we are sporadic, funded by ourselves and donations from individuals via Patreon starting from £1 (those who donate don't receive any extra access, it's just an opportunity for those who like the project to give some support). To keep up to date with Spryte's projects, please subscribe to our newsletter below!

Support Us! Sign Up!
Support Spryte with £1 Newsletter Sign-Up