00:02:28Model-wise double descent
00:04:39Sample-wise non-monotonicity
00:06:14Epoch-wise double descent
Lydfilen for denne episode er muligvis beskadiget
How to fix it?
OM DENNE EPISODE
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.
Engelsk
Det Forenede Kongerige
VIS NOTER 🔗
SØG BLANDT TIDLIGERE EPISODER
Søg efter gamle afsnit af AI Safety Fundamentals: Alignment.
ANDRE EPISODE I DENNE PODCAST
Chain-of-thought prompting has demonstrated remarkable performance on various natural language reasoning tasks. However, it tends to perform poorly on tasks which requires solving problems harder than the exemplars shown in the prompts. To overcome this challenge of easy-to-hard generalization, we …
(Sections 3.1-3.4, 6.1-6.2, and 7.1-7.5)
Suppose we someday build an Artificial General Intelligence algorithm using similar principles of learning and cognition as the human brain. How would we use such an algorithm safely?
I will argue that this is an open technical problem, and my goal in this …
This post is about language model scaling laws, specifically the laws derived in the DeepMind paper that introduced Chinchilla. The paper came out a few months ago, and has been discussed a lot, but some of its implications deserve more explicit notice in my opinion. In particular: Data, not size, …
In this post, we’ll present ARC’s approach to an open problem we think is central to aligning powerful machine learning (ML) systems:
Suppose we train a model to predict what the future will look like according to cameras and other sensors. We then use planning algorithms to find a sequence of ac…
Ansvarsfraskrivelse: Podcasten og kunstværket, der er indlejret på denne side, er fra BlueDot Impact, som tilhører dens ejer og ikke er tilknyttet eller godkendt af Listen Notes, Inc.
REDIG
Tak fordi du hjælper med at holde podcast-databasen opdateret.
