News for hackers. Stuff that matters.

A comment on How much did AlphaGo Zero cost? (2018).

Re:How much did AlphaGo Zero cost? (2018) by ipsum2 on Thursday June 11, 2020 @04:25AM (#23486432)
Alpha Go Zero*, which was trained from scratch, without human games.

I've also heard rumors that AlphaStar (https://deepmind.com/blog/article/alphastar-mastering-real-t...) was essentially put on hold because it was too expensive to improve/train. The bot wasn't able to beat StarCraft champions and _only_ got to a grandmaster level.

[ Parent | on HN ]
Re:How much did AlphaGo Zero cost? (2018) by Symmetry on Thursday June 11, 2020 @10:43AM (#23489434)
Alpha Go Zero used a combination of deep learning within a classical AI Monte Carlo simulation framework. Without a similarly effective framework its not surprising that Alpha Star wasn't able to achieve similar success. I've watched a lot of Alpha Star replays and the lapses in forethought glare through pretty regularly, though other aspects of its judgement (not just reflexes!) seem frankly super human.
[ Parent | on HN ]
Re:How much did AlphaGo Zero cost? (2018) by embrassingstuff on Thursday June 11, 2020 @06:05PM (#23493778)
Why was MCTS (or some search variant) not used in alphastar ?

(Sure, u need to somehow roll forward and rollback the StarCraft world, but for Atari using MCTS was shown to be an order of magnitude more efficient )

I have also seen comments that the search width is too large, or maybe academic purity consideration?

[ Parent | on HN ]
Re:How much did AlphaGo Zero cost? (2018) by __s on Thursday June 11, 2020 @12:21PM (#23490419)
https://www.youtube.com/watch?v=nbiVbd_CEIA

At the last Blizzcon they had it around. The setup wasn't ideal, so Serral (won world finals in 2018, reached semifinals in 2019) wasn't really happy with how he played, but it won

This was also a version where they'd worked on preventing its ability to micro at quadruple digit apm

[ Parent | on HN ]
Re:How much did AlphaGo Zero cost? (2018) by ajnin on Thursday June 11, 2020 @11:05PM (#23495601)
That match was still arguably not fair for the human because of the imperfect input and outputs he had to deal with. In the end I'm not sure it makes sense to put artificial handicaps on the machine to leave the human a chance. It's like a swimming race between a fish and some land animal but the fish is not allowed to use its fins. Sure the other guy has a better chance to win, but, what does that measure really?
[ Parent | on HN ]
Re:How much did AlphaGo Zero cost? (2018) by qayxc on Thursday June 11, 2020 @07:33AM (#23487695)
The main issue is that they need to retrain their bots for each new map, which would mean excessive training every time the map-pool changes (e.g. every 3 months or so IIRC).

I'd even argue that they missed their goal by a long shot if their system isn't able to play arbitrary maps - every human player can do that no problem.

[ Parent | on HN ]
Re:How much did AlphaGo Zero cost? (2018) by bsaul on Thursday June 11, 2020 @04:37AM (#23486504)
Interesting ! i didn't understand why they stopped the alphastar project. They pretended they reached their goal, but clearly haven't.
[ Parent | on HN ]
Re:How much did AlphaGo Zero cost? (2018) by NVHacker on Thursday June 11, 2020 @04:43AM (#23486548)
The goal was never to beat all pros. https://deepmind.com/blog/article/AlphaStar-Grandmaster-leve...
[ Parent | on HN ]
Re:How much did AlphaGo Zero cost? (2018) by sabas123 on Thursday June 11, 2020 @04:53AM (#23486604)
on a side note, I'm still disappointed to see that they didn't improve on the EAPM restriction to truly show they can beat humans without machine-y advantages.

Still what they have accomplished is a miracle.

[ Parent | on HN ]
Re:How much did AlphaGo Zero cost? (2018) by phonon on Thursday June 11, 2020 @06:56AM (#23487448)
The version that later went on ladder anonymously used human-like E/APM.

https://storage.googleapis.com/deepmind-media/research/alpha...

[ Parent | on HN ]
Re:How much did AlphaGo Zero cost? (2018) by empath75 on Thursday June 11, 2020 @04:31PM (#23493012)
playing at grandmaster level is a pretty astounding achievement.
[ Parent | on HN ]