Elon Musk's Open AI beat world best dota 2 players

A place to discuss anything.
Post Reply
cyxstudio
Posts: 297
Joined: Sun Mar 23, 2014 11:18 pm

https://www.youtube.com/watch?v=XbDmxEOj9OY

So basically they created real Artificial intelligence that plays against itself and learn stuff from scratch. This isnt the typical game AI where everything is pre scripted. They built the AI to learn by playing and accumulating lifetimes of experience.

Thinking about it, Is it possible to create an AI that can learn from betfair and trade betfair on pre race horse racing??

If they can do it for Chess, Go, and now Dota 2, i think its possible.
Last edited by cyxstudio on Wed Aug 16, 2017 3:43 pm, edited 1 time in total.
User avatar
Euler
Posts: 24701
Joined: Wed Nov 10, 2010 1:39 pm
Location: Bet Angel HQ

If you have a rigid environment with no external factors then you may be able to get something to learn it. But a lot of sport isn't like that.
User avatar
marksmeets302
Posts: 527
Joined: Thu Dec 10, 2009 4:37 pm

I've been thinking about this for some time already and am trying to learn the basic algorithms required for this challenge. A potential problem I see is that there's not enough data. With games like go and chess you can create new data by just playing a new game. With betfair, you need to wait for a new race to finish. Of course you can try the same races over and over again with different decisions but it doesn't change the environment as it does in a game where you have an opponent.

Thanks for the video!
User avatar
Euler
Posts: 24701
Joined: Wed Nov 10, 2010 1:39 pm
Location: Bet Angel HQ

The logic is harder to define in a real life sport IMHO. In most of the AI applications I have seen so far, it's working to rigid boundaries that can be tested and retested. How would AI know a horse is running loose? But also that it is on a circular track leading back to the start or open countryside?

Certain parts of the market are indefinable as well. If somebody is iceberging an order into the market that's a non-repeating pattern that's hard to pick up on but you can often easily see.

I'm sure it has applications, but only given the right criteria.
User avatar
ruthlessimon
Posts: 2094
Joined: Wed Mar 23, 2016 3:54 pm

Certainly pretty scary & people need to really think carefully about their futures. If they don't; they will get replaced by AI
User avatar
Euler
Posts: 24701
Joined: Wed Nov 10, 2010 1:39 pm
Location: Bet Angel HQ

User avatar
marksmeets302
Posts: 527
Joined: Thu Dec 10, 2009 4:37 pm

To be fair that's still better than most on a monday morning :D
User avatar
ruthlessimon
Posts: 2094
Joined: Wed Mar 23, 2016 3:54 pm

loved the first reply to that video - "looks like a real delivery worker to me" :lol: :lol:
poklius
Posts: 105
Joined: Sun May 17, 2009 11:58 am

In 1903 state of the art airplane hovered for 12 seconds and covered 120 feet. 66 years later we landed in moon. Exponential technology growth can be freakin scary. I dont see any reason why AI couldn't trade markets or "watch" soccer match and bet inplay, all the technology needed is here. Just some time and fine tuning is needed..
Wainwright
Posts: 108
Joined: Mon Jan 23, 2012 1:10 pm

Stratagem, the startup mentioned in a recent thread, are apparently doing precisely that;
Stratagem’s AI makes its calculations watching a standard, broadcast feed of the match. (Pro: it’s readily accessible. Con: it has to learn not to analyze the replays.) It tracks the ball and the players, identifying which team they’re on based on the color of their kits. The lines of the pitch are also highlighted, and all this data is transformed into a 2D map of the whole game. From this viewpoint, the software studies matches like an armchair general: it identifies what it thinks are goal-scoring chances, or the moments where the configuration of players looks right for someone to take a shot and score.
https://www.theverge.com/2017/7/6/15923 ... -stratagem

I don't see it as much of a threat to traditional trading though as they are training it by showing it historical games - once it gets to LVGs reign at Man Utd it will probably pack up through sheer boredom.
User avatar
Euler
Posts: 24701
Joined: Wed Nov 10, 2010 1:39 pm
Location: Bet Angel HQ

Never confuse marketing with actual application in the market.

Stratagem feels like Centaur 2 to me.

If you have a brand new technology that nobody has, you just unleash it to the market and clean up.
Wainwright
Posts: 108
Joined: Mon Jan 23, 2012 1:10 pm

I imagine it to be something like this;

https://www.youtube.com/watch?v=ACmydtFDTGs
User avatar
workpeter
Posts: 165
Joined: Sat Jul 30, 2016 8:29 pm

It would be very challenging because unlike games there is less consistency. If I were to attempt using the same approach I would first need the ability to capture and replay a stream so the AI can replay the same race over and over using random strategies. This, of course, would need to be done across hundreds of races and indefinitely in order to keep updated with ever changing markets.
User avatar
Euler
Posts: 24701
Joined: Wed Nov 10, 2010 1:39 pm
Location: Bet Angel HQ

I looked at training neural nets back in 2007, but it only seems to be trendy now. I'm not sure what has changed, computing power? The methodologies don't seem radically different?
Post Reply

Return to “General discussion”