Episode 33: Tri Dao, Stanford: On FlashAttention and sparsity, quantization, and efficient inference
Generally Intelligent
The Theory of Modern Data Augmentation
In 2018-2019 we saw a lot of computer vision models that improve the accuracy, mostly by augmenting data. We want to understand what people have been doing in practice and kind of shine on more theoretical light into what that works. It's part of a larger trend of not just looking at the model design, but also looking at the data that the models are trained on.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
