@MakunaHatata@lemmy.ml to Lemmy Shitpost@lemmy.world • 4 months agoAI or DEI?lemmy.mlimagemessage-square103fedilinkarrow-up1323
arrow-up1323imageAI or DEI?lemmy.ml@MakunaHatata@lemmy.ml to Lemmy Shitpost@lemmy.world • 4 months agomessage-square103fedilink
minus-square@bamboolinkEnglish10•4 months agoRather than eliminating the some of the training data, you could add more training data to create an even balance.
minus-square@kromem@lemmy.worldlinkfedilinkEnglish3•4 months agoIndeed - there’s a very good argument for using synthetic data to introduce diversity as long as you can avoid model collapse.
Rather than eliminating the some of the training data, you could add more training data to create an even balance.
Indeed - there’s a very good argument for using synthetic data to introduce diversity as long as you can avoid model collapse.