Ideas & society2020 · shared on Hacker News

Every Model Learned by Gradient Descent Is Approximately a Kernel Machine

Explore Every Model Learned by Gradient Descent Is Approximately a Kernel Machine, its original source and Hacker News discussions. Discover community perspectives and save it to your reading list.

Read original

The conversation around the idea.

Original Hacker News threads. Different perspectives, questions, and things worth thinking about.

Showing up to 100 related threads. A paper may appear in a comment rather than the original submission.

Links found in the discussion

Up to 50 recent source links, in HN item order.

Yscottlocklin ·

Every Model Learned by Gradient Descent Is Approximately a Kernel Machine

406 points107 comments
Yaflip ·

Ask HN: What things are happening in ML that we can't hear over the din of LLMs?

364 points99 comments
Yumangkeshri ·

Does GPT-2 Know Your Phone Number?

321 points155 comments
Yagnosticmantis ·

Deep learning is hitting a wall?

274 points250 comments
Yhomarp ·

Pathways Language Model (PaLM): Scaling to 540B parameters

235 points207 comments
YAnon84 ·

Every model learned by gradient descent is approximately a kernel machine (2020)

176 points136 comments
Ythe__prestige ·

A new link to an old model could crack the mystery of deep learning

173 points66 comments
Ytibbar ·

Using GPT-3 to pathfind in random graphs

165 points79 comments
Y1wheel ·

Do Machine Learning Models Memorize or Generalize?

7 points0 comments
Ybreck ·

Every Model Learned by Gradient Descent Is Approximately a Kernel Machine

4 points0 comments
YLordNibbler ·

Every Model Learned by Gradient Descent Is Approximately a Kernel Machine

4 points0 comments
Ydeathflute ·

Deep learning networks are approximately kernel machines

2 points1 comments
Yoptimalsolver ·

Every Model Learned By Gradient Descent Is Approximately A Kernel Machine (2020)

2 points0 comments
Ygardenfelder ·

Google researchers deal a major blow to the theory AI about to outsmart humans

2 points1 comments
Ycgadski ·

When Gradient Descent Is a Kernel Method

1 points0 comments

One idea leads to another.

Explore more papers