orbifold

34 ms·

361.
by orbifold 6y ago
That is a nice article! Thanks for pointing it out. Most of the time I try to bite my tongue aswell. It is just sometimes really upsetting.
362.
by orbifold 6y ago
There are few other ways of probing high energy physics systematically and a lot of them are funded and done aswell. This is not a zero sum game. We can bury detectors in the antarctic, in mines and have CERN. In fact technology developed a
363.
by orbifold 6y ago
An organisation like CERN is extremely hard to replicate and easily destroyed and dragged down by a thousand cuts. The EU is funding similar moonshots in Neuroscience , Graphene, Quantum Computing and most recently machine learning. They al
364.
by orbifold 6y ago
Building a high energy collider is like building a larger telescope. The only things you can be sure of is that you will be able to understand whatever you have seen better than before. There is scientific consensus that we should be able t
365.
by orbifold 6y ago
That was uncalled for, so I took it out.
366.
by orbifold 6y ago
Science shouldn’t be a zero sum game. As a physicist I love that we get a huge chunk of science funding for all kinds of stuff. Generally speaking developing a new accelerator leads to excellent technical advances at all levels of the stac
367.
by orbifold 6y ago
This is a bit of an oversimplification, because I don't actually know the details, but it works roughly like this: GPUs are devices on the PCI Express bus, without an operating system getting in the way everything on the PCI Express bu
368.
by orbifold 6y ago
A good physicist has the ability to grasp a large fraction of all physics that has ever been done. And there is a pretty good directed knowledge graph. The action principle and some ideas from quantum mechanics are enough to anchor almost a
369.
by orbifold 6y ago
It is not ideal but for the subset of libraries that use CMake it can be as easy as dumping the library in a subdirectory and you are done. Things only start to go sideways when you have to integrate with google libraries since they have ty
370.
by orbifold 6y ago
Lattice QCD is something legitimate that people have been working on since the 80s. Some of the earliest massively parallel machines like the Connection Machine were thought to be useful for it. It’s totally plausible that physicists were c
371.
by orbifold 6y ago
Wouldn’t be the end of the world. Most of those are really toxic and in the pocket of the NSA. Content licensing could lead to local competitors of Netflix easily.
372.
by orbifold 6y ago
It seems very unlikely that there can be vaccine that is effective long term. Just as there is no long term Flu vaccine.
373.
by orbifold 6y ago
Doing new drivers is not a huge deal. At least you are programming them in C with sane tooling against a sane API. Getting up a simple setup with a bit of DMA and a few registers is not that hard. It pales compared to the amount of pain yo
374.
by orbifold 6y ago
time to precompile Plots: 138s julia> @time Plots.scatter(data.eq_site_limit, data.hu_site_limit) 5.461595 seconds subsequent calls report significant speed ups compared to matplotlib, but render about as quickly julia> @time Plots.
375.
by orbifold 6y ago
So I want to plot the data I just loaded. Since I wrote the comment I had closed the Julia window (not to make a point, btw.) julia> import CSV (takes ~1-2 seconds) julia> data = CSV.File("/User/orbifold/Downloads&
376.
by orbifold 6y ago
I installed CSV.jl and tried it out on the insurance dataset from here " https://support.spatialkey.com/spatialkey-sample-csv-data/ ". First import took >10s just to precompile and then CSV.File took 5.21s t
377.
by orbifold 6y ago
You don't have to mandate something that people have done voluntarily. Mobile phones are cameras and microphones in every house, school and workplace. All the mobile standards have "legal intercept" provisions, which basicall
378.
by orbifold 6y ago
It should be "Scholze" of course.
379.
by orbifold 6y ago
I might misremember it slightly, I got told this by a professor (he was one of his examiners) and people that knew him at Bonn in ~2010. He could also have been awarded the majority of the credit points towards the Bachelor at entrance.
380.
by orbifold 6y ago
The starkest example to me of breadth first vs depth first learning are math geniuses. They typically ignore everything except mathematics to begin with and then hyper focus on a specific area of mathematics intensely. The most striking I k
381.
by orbifold 6y ago
Or they have a secretary that does ;)
382.
by orbifold 6y ago
Here is my main gripe with Julia: They blatantly copy Matlab. The same is also true to a lesser degree of Python's numpy and matplotlib, but Julia goes the whole nine yards, by basically replicating the syntax and copying most of the n
383.
by orbifold 6y ago
As a scientist: The amount I struggled with getting distributed training on a HPC cluster to work vs. how easy it was with Lightning was eye opening. Almost no code change and finally I can run across 20 nodes with 4 V-100 each :). Plus the
384.
by orbifold 6y ago
Might not be your target audience, but high energy physics has been operating an infrastructure like that for years: https://www.etp.physik.uni-muenchen.de/research/grid-computi... You can basically use these framewor
385.
by orbifold 6y ago
It doesn’t apply to the universe because of GR at least that is the conventional view I think.
386.
by orbifold 6y ago
Exactly, it encourages mafia like structures. It matters a ton who you know and who your PI was as a Postdoc and PhD.
387.
by orbifold 6y ago
They don't address the fundamental issue of BPTT in an RNN model. To my mind this is a much larger block to biological plausibility than how to track the error locally. Besides this has been addressed in various forms in earlier work:
388.
by orbifold 6y ago
All the major toolchains work on Linux. I personally liked Pynq http://www.pynq.io/ , it works fairly well and is beefy enough to do non-toy projects with it. For something totally barebones you can try https://ww
389.
by orbifold 6y ago
You need at least 10x over an established approach to have any hope of surviving in the mid-term. The results are for batch-size 1 as well. GPUs are throughput optimized. So I wouldn't be surprised if an A100 chip would be able to outp
390.
by orbifold 6y ago
Research in this area has always been cyclical. The Neural ODE paper that won a best paper award at Neurips two years ago just implements a technique that is common knowledge in the optimal control theory community since the 60s. I wouldn’t

More ›