Horizon: Facebook's Open Source Applied Reinforcement Learning Platform — mentioned by agbell

Mon, 01 Feb 2021 18:23:51 GMT

Thanks! My understanding is they are looking at page management activity and whether it increases above what they would expect if they didn't notify. Some of the details are covered in the paper [1] >The Markov Decision Process (MDP) is based on a sequence of notification candidates for a particular person. The actions here are sending and dropping the notification, and the state describes a set of features about the person and the notification candidate. There are rewards for interactions and activity on Facebook, with a penalty for sending the notification to control the volume of…

Paper and discussion history. A mention is not an endorsement.

Horizon: Facebook's Open Source Applied Reinforcement Learning Platform — mentioned by xiphias2

Tue, 27 Nov 2018 20:17:06 GMT

The paper is here: https://arxiv.org/pdf/1811.00260.pdf It's interesting that it's already used for optimizing for long term value when deciding whether a push notification should be sent or not.

Paper and discussion history. A mention is not an endorsement.

Horizon: Facebook's Open Source Applied Reinforcement Learning Platform · HackerPapers discussions

Horizon: Facebook's Open Source Applied Reinforcement Learning Platform — mentioned by agbell

Horizon: Facebook's Open Source Applied Reinforcement Learning Platform — mentioned by xiphias2