Horizon: Facebook’s Open Source Applied Reinforcement Learning Platform

#1

Post by inputcoffee » Thu, Nov 01, 2018, 6:18 PM UTC

Horizon: Facebook’s Open Source Applied Reinforcement Learning Platform
research.fb.com

Re: Horizon: Facebook’s Open Source Applied Reinforcement Learning Platform

#2

Post by inputcoffee » Thu, Nov 01, 2018, 6:20 PM UTC

In case you want to see the actual paper: https://research.fb.com/wp-content/uploads/2018/10/Horizon-F...?

Re: Horizon: Facebook’s Open Source Applied Reinforcement Learning Platform

#3

Post by rjammala » Thu, Nov 01, 2018, 6:47 PM UTC

Github repo: https://github.com/facebookresearch/Horizon

Re: Horizon: Facebook’s Open Source Applied Reinforcement Learning Platform

#4

Post by amrrs » Thu, Nov 01, 2018, 6:47 PM UTC

Discussion on Google's Dopamine - its Reinforcement Learning Framework https://news.ycombinator.com/item?id=15648746

Re: Horizon: Facebook’s Open Source Applied Reinforcement Learning Platform

#5

Post by traek » Thu, Nov 01, 2018, 7:24 PM UTC

amrrs wrote: post #4

Discussion on Google's Dopamine - its Reinforcement Learning Framework https://news.ycombinator.com/item?id=15648746

That discussion is on Dopamine the startup, not the framework from Google.

Google's framework is at https://github.com/google/dopamine and I don't believe it's generated discussion on HN before.

Re: Horizon: Facebook’s Open Source Applied Reinforcement Learning Platform

#6

Post by maldeh » Thu, Nov 01, 2018, 7:57 PM UTC

rjammala wrote: post #3

Github repo: https://github.com/facebookresearch/Horizon

Ah, it has the fabled BSD+ license: https://github.com/facebookresearch/Horizon/blob/master/PATE...

I guess that's not too unusual from their opensource contributions at this point.

Re: Horizon: Facebook’s Open Source Applied Reinforcement Learning Platform

#7

Post by Diederich » Thu, Nov 01, 2018, 9:39 PM UTC

inputcoffee wrote: post #2

In case you want to see the actual paper: https://research.fb.com/wp-content/uploads/2018/10/Horizon-F... ?

Thanks for the link!

I recommend that anybody who is interested in this read section 9 in this paper: "NOTIFICATIONS AT FACEBOOK". It brings into focus real ways that this technology is used and is useful.

Re: Horizon: Facebook’s Open Source Applied Reinforcement Learning Platform

#8

Post by pesenti » Thu, Nov 01, 2018, 9:46 PM UTC

Blog post: https://code.fb.com/ml-applications/horizon/ (would be a better link if that can be changed)

Re: Horizon: Facebook’s Open Source Applied Reinforcement Learning Platform

#9

Post by Geee » Thu, Nov 01, 2018, 11:00 PM UTC

This kind of AI applied at scale (Facebook) scares me. Mainly because humans are at the other side of the feedback loop, and not only is the AI adapting, but people are adapting too. Over time this kind of runaway feedback loop could lead into anything.

Re: Horizon: Facebook’s Open Source Applied Reinforcement Learning Platform

#10

Post by sova » Thu, Nov 01, 2018, 11:02 PM UTC

Diederich wrote: post #7

inputcoffee wrote: post #2

In case you want to see the actual paper: https://research.fb.com/wp-content/uploads/2018/10/Horizon-F... ?

Thanks for the link! I recommend that anybody who is interested in this read section 9 in this paper: "NOTIFICATIONS AT FACEBOOK". It brings into focus real ways that this technology is used and is useful.

So cryptic and hand wavy description in paper but hey I'm glad they have good results. Going from pedals to combustion engines clearly has efficiency benefits but they gloss over the real flesh of the fruit when they don't describe how their particular engine is better or tuned. Wish they had more talk of notifications as Markov chains and how they evaluate interaction