It's exactly at the level I thought was communicated by the title and introduction.
"Human level competitive", "solidly amateur-level human performance", "beat 100% of beginners and 55% of intermediate players". That robot would definitely win some games in your local club league, except that it doesn't serve, and unless it's cheating in ways the announcement glosses over like extra cameras - DeepMind have some history here so I reserve the right to be skeptical.
The only thing I'd take issue with in the abstract is "Table tennis... requires human players to undergo years of training to achieve an advanced level of proficiency." While that sentence is true, it's irrelevant to this robot since this robot only plays at intermediate proficiency, a level reachable by a moderately athletic human with some practice.
By contrast, the AlphaGo [0] AlphaZero [1] and AlphaStar [2] papers claim "mastery", "superhuman", "world champion level", "Grandmaster-level", "human professional" ability - all defensible claims given their performance and match conditions in the respective games.
> That robot would definitely win some games in your local club league
Definitely not. If you go beyond the cherry-picked videos where some longer sequences happened, the longer match videos reveal how bad the robot is. It makes really bad mistakes and loses most points against players not even on intermediate player in any local club.
Yeah, looking it play makes me thinks it has a level comparable to mine, that is the level of someone that enjoys playing table tennis with friends and family members a every other year. Not at all the level you'd see in clubs, even among low-ranked people.
I'd describe the robot's level as "good for outdoor table tennis".
There are pretty much too distinct classes of players. Those that occasionally play for fun, typically at stone tables found in parks and open-air baths.
And then there are those that play and train at least once a week in indoor halls with wooden tables, and often try to learn proper stroke techniques, and often participate in leagues.
The robot is pretty good fit for the first category, and that's already a pretty impressive achievement.
In the second category, it'd lose to anybody playing for more than a year or two, so it would be on par with the lowest tier players there.
My local club is full of (mostly older) players who play once a week for a few hours. They don't train outside of this, but they play with proper equipment, sometimes play leagues, sometimes play a few strokes well, and beat anyone who just plays casually, apart from the occasional excellent tennis player who can transfer just enough skills to be competitive. But they also make lots of unforced errors and don't have good technique on all strokes.
The robot looks like it would be competitive with most of those players. Maybe my club is uniquely weak.
The AlphaZero paper claimed it was vastly superior to Stockfish, the top open source engine, based on a 100-game match against Stockfish. It turned out Stockfish was running with way less compute. It's not an apples to apples comparison between CPU and GPU, but IIRC there were orders of magnitude difference in the hardware cost and power budget. Additionally, they used a build of Stockfish that was tuned based on having access to opening books and endgame tablebases, but didn't give it those resources in the match.
The original AlphaStar announcement was also based on having serious advantages over its human opponents: it got a feed of the whole map, where humans could only view a section at a time, and the ability to perform an unrealistic number of actions per minute.
The equivalent in table tennis? Maybe having an additional high speed camera on the other side of the table, or a sensor in the opponent's bat. Actually, why is the opponent playing with a non-standard bat with two black rubbers? Presumably that's an optimization where the robot's computer vision has been tuned only for a black bat. But if that's so, it means none of the opponents got to use their own equipment, they used a bat which was unfamiliar and perhaps chosen to be easy for the robot to play against.
I could concede it as "lower intermediate", but the span is huge. I don't think it can beat an upper beginner consistently. It all depends on what ELO you consider intermediate.
"Human level competitive", "solidly amateur-level human performance", "beat 100% of beginners and 55% of intermediate players". That robot would definitely win some games in your local club league, except that it doesn't serve, and unless it's cheating in ways the announcement glosses over like extra cameras - DeepMind have some history here so I reserve the right to be skeptical.
The only thing I'd take issue with in the abstract is "Table tennis... requires human players to undergo years of training to achieve an advanced level of proficiency." While that sentence is true, it's irrelevant to this robot since this robot only plays at intermediate proficiency, a level reachable by a moderately athletic human with some practice.
By contrast, the AlphaGo [0] AlphaZero [1] and AlphaStar [2] papers claim "mastery", "superhuman", "world champion level", "Grandmaster-level", "human professional" ability - all defensible claims given their performance and match conditions in the respective games.
[0] https://www.researchgate.net/publication/292074166_Mastering...
[1] https://arxiv.org/pdf/1712.01815
[2] https://www.nature.com/articles/s41586-019-1724-z