hah maybe. I feel like I would start with TransAct, and maybe PinSage to cover the world of recs at Pintrest

also a bunch of practical stuff on how they serve these long user histories. Clearly a lot of work went into this, and a lot of goodies in this paper.

- ranking model that tried to predict the next user action in a sequence. Use a cross-entropy loss to predict labels/actions on the sample and a next action loss to predict the next action the user will take.
- impression based negative sampling doubled some offline metrics

but some more than others

you most of all

This person has clearly never used protobufs, especially in python.

Hah a friend and I managed to get something out of it. Though mostly Cronenberg isn’t ok. And I guess tech enables unhealthy obsessions

love your work, fyi some text got dropped. First paragraph of 2. ends with "Saka’s proposal would have reduced" and then it goes onto the next paragraph.

There was an ML involved reboot. Investigations are ongoing.

I just biked through that very park, less snow today! Any idea why all the lights are off at night time? Is spooky.