Mike

143 posts

Mike

Mike

@MikeJInnes

Edinburgh Katılım Ekim 2011
146 Takip Edilen793 Takipçiler
Mike
Mike@MikeJInnes·
@jeremyphoward Well now, someone's co-founder is sleeping in the spare room tonight
English
0
0
0
143
Jeremy Howard
Jeremy Howard@jeremyphoward·
Yay Flash 2.5 knows of FastHTML! :) It's not yet very good at it without providing llms.txt context, but it's a start. (Also, who is Mike Innes?...)
Jeremy Howard tweet media
English
12
3
127
11.7K
Mike
Mike@MikeJInnes·
@HeblingVieira Yeah, next I need to try making ChatGPT the player!
English
0
0
1
0
Mike
Mike@MikeJInnes·
Did not expect to be role-playing with a GPU this evening.
Mike tweet media
English
3
1
14
0
Mike
Mike@MikeJInnes·
Mike tweet mediaMike tweet mediaMike tweet mediaMike tweet media
ZXX
0
0
2
0
Mike
Mike@MikeJInnes·
@oxinabox_frames @apaszke @shoyer @dancherp On this view you get dual numbers by implementing an instruction tape tracer + transform, and then optimising it by fusing those steps together, without changing the semantics or PC issues.
English
0
0
2
0
Mike
Mike@MikeJInnes·
@oxinabox_frames @apaszke @shoyer @dancherp I tend to think of the tracing and AD transform steps as orthogonal. AD transform looks similar everywhere (adjusted for Fortran AST, Jaxprs or SSA). Tracing systems are the source of PC errors, not the transform step.
English
1
0
1
0
Frames Catherine White
Frames Catherine White@oxinabox_frames·
Can someone (@dancherp ?) talk to me (or point me to a writeup) about the historical decision to move the Swift AD work from having distinct co-tangent and tangent types, to just having common tangent type used for both forward and reverse? I want to understand pros and cons.
English
1
0
11
0
Mike
Mike@MikeJInnes·
"The police and the national guard, who ringed every campus, keeping the students from creeping across to society like so many black rats swarming out of a leaky ship.”
English
0
0
1
0
Mike
Mike@MikeJInnes·
“He knew that at one time she had been illegally married to a student commune leader, and that for one year she had lived in the rabbit warrens of Columbia University, along with all the smelly, bearded students kept subsurface lifelong by the pols and the nats. ...
English
1
0
0
0
Mike
Mike@MikeJInnes·
“I can’t live two hours without my ID, he said to himself. I don’t even dare walk out of the lobby of this rundown hotel and onto the public sidewalk. They’ll assume I’m a student or teacher escaped from one of the campuses.” – Philip K. Dick
English
1
0
0
0
Mike
Mike@MikeJInnes·
@vflorelo Yup, this is the source!
English
0
0
1
0
Mike retweetledi
John Carmack
John Carmack@ID_AA_Carmack·
The Imperial College epidemic simulation code that I helped a little on is now public: github.com/mrc-ide/covid-… I am a strong proponent of public code for models that may influence policy, and while this is a "release" rather than a "live" depot, it is a Good Thing.
English
19
395
1.5K
0
Mike
Mike@MikeJInnes·
@skornblith @StefanKarpinski The SK / Singapore approach depends on pretty extreme measures (location tracking, loose rules on gov access to medical records etc) that will be a lot harder to implement in the US (politically and logistically) than lockdown
English
0
0
1
0
Simon Kornblith
Simon Kornblith@skornblith·
@StefanKarpinski After 2 months of lockdown, hopefully the disease will not be as widespread as it is now. At that point, we could test all close contacts of anyone who tests positive, which might be enough. This seems to have worked in South Korea; not yet clear if it will work in the U.S.
Simon Kornblith tweet media
English
2
0
3
0
Mike
Mike@MikeJInnes·
@StefanKarpinski A lot depends on how long immunity lasts for; for some coronaviruses it's around three months, which would make things significantly harder. Hopefully more data coming out on that soon.
English
1
0
3
0
Mike
Mike@MikeJInnes·
Imperial College paper that has informed UK pandemic policy. Well-explained exploration of some of the tradeoffs involved imperial.ac.uk/media/imperial…
English
0
3
3
0
Mike
Mike@MikeJInnes·
@ChadScherrer @theshawwn @rbhar90 That's not the only option; e.g. you can just make up a derivative that's good enough for gradient descent. And there are probably other mathematically reasonable schemes that you could choose.
English
0
0
3
0
Mike
Mike@MikeJInnes·
@ChadScherrer @theshawwn @rbhar90 All programs have a bunch of non-differentiable operations like mod. An AD will just assume the derivative is always 0 for that function.
English
3
0
1
0
Mike
Mike@MikeJInnes·
@Inoryy @ianadwilliamson @shoyer Julia's Python interop is actually really solid, and in some ways more convenient (no PyObject casting, more auto conversions). More experimental (but basically working) is differentiation through PyCall: #slide=id.g5ba58bae6c_0_115" target="_blank" rel="nofollow noopener">docs.google.com/presentation/d…
English
1
0
9
0
Roman Ring
Roman Ring@Inoryy·
@ianadwilliamson @shoyer To be clear, I know that Python interop exists to some extent in Julia s well via PyCall, but from what I've gathered it's more of an experimental feature developed by a third-party. Would be happy to learn more if I'm mistaken.
English
1
0
0
0
Roman Ring
Roman Ring@Inoryy·
Have you ever wondered what will be the ML frameworks of the '20s? In this essay, I examine the directions AI research might take and the requirements they impose, concluding with an overview of what I believe to be two strong candidates: JAX and S4TF. inoryy.com/post/next-gen-…
English
9
116
471
0
Mike
Mike@MikeJInnes·
@ChadScherrer @oxinabox_frames A big challenge is getting efficient source transform AD and higher-order derivatives; if done naively, nesting gradients produces exponential amounts of code. You'll notice this if you try to do third-order in Zygote
English
0
0
5
0
Frames Catherine White
Frames Catherine White@oxinabox_frames·
Seems like a lot of ML/Deep Learning folk think that automatic differentiation is a solved problem. This is very similar to NLP or Computer Vision people thinking that ML is a solved problem.
English
1
0
14
0