))
Deep Double Descent
Manage episode 424087967 series 3498845
Contenuto fornito da BlueDot Impact. Tutti i contenuti dei podcast, inclusi episodi, grafica e descrizioni dei podcast, vengono caricati e forniti direttamente da BlueDot Impact o dal partner della piattaforma podcast. Se ritieni che qualcuno stia utilizzando la tua opera protetta da copyright senza la tua autorizzazione, puoi seguire la procedura descritta qui https://it.player.fm/legal.
We show that the double descent phenomenon occurs in CNNs, ResNets, and transformers: performance first improves, then gets worse, and then improves again with increasing model size, data size, or training time. This effect is often avoided through careful regularization. While this behavior appears to be fairly universal, we don’t yet fully understand why it happens, and view further study of this phenomenon as an important research direction.
Source:
https://openai.com/research/deep-double-descent
Narrated for AI Safety Fundamentals by Perrin Walker of TYPE III AUDIO.
---
A podcast by BlueDot Impact.
Learn more on the AI Safety Fundamentals website.
Capitoli
1. Deep Double Descent (00:00:00)
2. Model-wise double descent (00:02:28)
3. Sample-wise non-monotonicity (00:04:39)
4. Epoch-wise double descent (00:06:14)
83 episodi
Condividi
