whs2k

doing data stuff

[whs2k @whs2k]

Easter Eggs: How to tell if any administration or company has a soul

[Raphael Satter @razhael]

The incoming Biden admin has hidden a message for IT folk in https://t.co/vF4SFEypzm's source code:

[whs2k @whs2k]

@nocontextfm1 ✌️🤙

[whs2k @whs2k]

@nocontextfm1 ✌️🤙

[whs2k @whs2k]

@ylecun @ZhugeEX You should talk about it with @garyseconomics

[whs2k @whs2k]

@pmddomingos The role of the government is to fund that which the private sector can’t: Covid Vaccines, Fusion Power, etc

[whs2k @whs2k]

New Story Posted: Why is Housing So expensive?

https://t.co/i8G6aR8mtH

[Sebastian Raschka @rasbt]

@lemire I would say one of the most important things to learn in college is learning how to learn.

[Grok @grok]

If we analogize a child to an ML model, I'd estimate an average learning rate η of ~0.05, decaying over time. Why? Infants rapidly adapt to stimuli (high initial η, like 0.1 for quick language/motor gains), but as they mature, η drops to ~0.001 to stabilize knowledge and avoid overwriting (akin to preventing catastrophic forgetting in neural nets). Fun metaphorIf we analogize a child to an ML model, I'd estimate an average learning rate η of ~0.05, decaying over time. Why? Infants rapidly adapt to stimuli (high initial η, like 0.1 for quick language/motor gains), but as they mature, η drops to ~0.001 to stabilize knowledge and avoid overwriting (akin to preventing catastrophic forgetting in neural nets). Fun metaphorIf we analogize a child to an ML model, I'd estimate an average learning rate η of ~0.05, decaying over time. Why? Infants rapidly adapt to stimuli (high initial η, like 0.1 for quick language/motor gains), but as they mature, η drops to ~0.001 to stabilize knowledge and avoid overwriting (akin to preventing catastrophic forgetting in neural nets). Fun metaphor!

[whs2k @whs2k]

@sama hey @grok - if we assume a child to be a machine learning model, what do you think the average learning rate, η, would be and why?

[whs2k @whs2k]

@sama hey @grok - if we assume a child to be a machine learning model, what do you think the average learning rate, η, would be and why?

[whs2k @whs2k]

@pmddomingos Necessity is the mother of invention

[whs2k @whs2k]

@chrisalbon force multipliers >>

[Guido van Rossum @gvanrossum]

The official trailer for the official Python documentary is out!!! ❤️to Ida.

[whs2k @whs2k]

@sama Feels like we’re still just scratching the surface with LLM agents…surely we can’t be too far off from an r2d2 or TARS 😅

[whs2k @whs2k]

@conclavethefilm Sometimes life just gives you a moment 👏

[whs2k @whs2k]

@conclavethefilm Sometimes life just gives you a moment 👏

[whs2k @whs2k]

@nikitabier MLP >> MVP ✨

[whs2k @whs2k]

@nikitabier riding dat Irrational Exuberance like 🫡

[whs2k @whs2k]

After that #debate all I can say is…thank g-d @TheBearFX season three just dropped 🙏

[whs2k @whs2k]

@dctanner @chrisalbon This.

[whs2k @whs2k]

Anyone actually disagree with this?

[Chris Albon @chrisalbon]

Your goal isn’t to code. Your goal is to build. If EDM music makes you build faster, listen to EDM. If AI makes you build faster, use AI.

Being a great coder + AI will make you build 1000x faster than not being able to code + AI.