Cast by jc4p

Show the cast this replies to

imo the root issue is all their algorithms are supervised (they label and tag things and give them scores and adjust) vs what everyone else has been doing for 10+ years which is unsupervised (the system learns the labels and rules) — the current approach 1. doesn’t scale 2. doesn’t work with other languages

12 likes0 recasts3 replies

3 replies

  • that current approach seems very…uh…arrogant(?) in that it’s so subjective “WE can most accurately determine scores and weighting”

    8 likes · 0 recastsRead 2 replies

    💯💯💯 it is great for a young network but a chokehold now, + it also has a large sense of control involved (“_I_ get to pick what’s viral on _my_ platform”) but it just doesn’t scale at all

    6 likes · 0 recastsRead 1 reply

    maybe this is what it means to be a sufficiently decentralised social media platform, but yes the algorithm sucks and no amount of onboarding can help us if we fail to retain creators again and again

    6 likes · 0 recastsRead 1 reply

    most

    No likes or recasts yet
  • Exactly. Not able to get personalization to cater feeds this way

    No likes or recasts yet
  • I think you're conflating supervised vs. unsupervised training with manual vs. learned models. We used supervised training, because the model needs feedback from users to know what they like. We don't use manual heuristics, but instead train models that infer what users like from the feedback they're given. This does scale because they feedback is generated when users click, like or otherwise engage with content. It is not generated by us manually labeling things.

    1 like · 0 recastsRead 1 reply

    appreciate the response and totally acknowledge the oversimplification! my issue (which i have no way of knowing it's a wild guess) is that the models are trained on global (and possibly handpicked) features, the vibe i get is there isn't clustering and then each cluster gets its own features

    No likes or recasts yetRead 1 reply

    the features are personalized to each user (so not global) but are hand-picked. The model does some implicit clustering based on the features we feed it, but we don't have explicit clustering of topics or users today. Something we're thinking about actively rn.

    3 likes · 0 recastsRead 1 reply