currymj

71 ms·

841.
by currymj 8y ago
there already is one for Visual Studio! https://www.microsoft.com/en-us/quantum/development-kit It's actually a reasonable way to play around with simulations of quantum algorithms, much nicer than multiplyin
842.
by currymj 8y ago
elsewhere it suggests “Python + Numba” and “Python + C” as possibilities. i have to assume that numpy would be okay. i mean, it’s close to being part of the standard library at this point.
843.
by currymj 8y ago
there's a nonzero amount of genuine AI/machine learning going on in the research end of things right now in this field. but you're right, as is currently deployed, a better title would be "how mathematical optimization c
844.
by currymj 8y ago
Oh and I don't think you'll be as hopeless a coder as some other posters might think, because you did your PhD in a computational discipline and know Python and have heard of unit testing. There are, say, physics PhDs who only wri
845.
by currymj 8y ago
I have made the jump from writing academic code to working on a product where actual software engineering was encouraged. Although I did jump back to academia pretty quickly. Hitchhiker's Guide to Python is a very good book (freely ava
846.
by currymj 8y ago
it is considered a perk. good tech jobs, for instance, often have paid maternity leave, while cleaning and food contractors at these companies might not get this. since up to 12 unpaid weeks is mandatory, people find a way to make it work,
847.
by currymj 8y ago
this is a stereotype of California, not just San Francisco. people make the same observation about film industry types. like every seemingly irrational and inefficient mode of cultural interaction, there are some benefits to it.
848.
by currymj 8y ago
One of the core equations in reinforcement learning is the Bellman equation -- named after Richard Bellman, inventor of dynamic programming. And in fact, in the operations research community, reinforcement learning is often referred to as &
849.
by currymj 8y ago
The recent game EXAPUNKS is definitely like this, and worth playing, although it's assembly language programming puzzles so might be a sort of busman's holiday for a lot of HN readers.
850.
by currymj 8y ago
but not anonymous — they are definitely monitored, not really useful for insiders to do insider trading. after all, Matt Levine’s second rule of insider trading is “don’t do it by buying short-dated out-of-the-money call options on merger t
851.
by currymj 8y ago
Your average person even in the US feels that there is something unsavory about short selling, like they are saboteurs, gleefully profiting off destruction. Even people you'd expect to be more sophisticated feel this way; I don't
852.
by currymj 8y ago
well, is it legal to do a "clean-room" implementation from the descriptions in the papers themselves, without looking at any provided code? (this should almost always be feasible, and is commonly done for non-IP-related reasons e.
853.
by currymj 8y ago
shellcode, maybe, or really tight inner loops that need to use weird vector instructions.
854.
by currymj 8y ago
same as it’s legal to use open source licensed software to profit. a lot of academics don’t care, and their work might be built very strongly on other work. there might or might not be a patent on this research, but I would guess probably n
855.
by currymj 8y ago
it would be expensive in terms of hardware, you’d need to shell out for quite a few GPUs or else budget a lot for expensive cloud instances. plus this is just a very complicated thing, in that it’s gluing together multiple new techniques to
856.
by currymj 8y ago
you may enjoy this article: http://slatestarcodex.com/2018/07/10/melatonin-much-more-tha... if you scroll towards the bottom it talks about how you can use melatonin to deal with a weird sleep cycle like this
857.
by currymj 8y ago
well, the Language Server Protocol was developed originally for use in VS Code. So not surprising it works better! (i also wish lsp-mode would improve.)
858.
by currymj 8y ago
there is definitely a naive perception in some circles right now that a doctorate studying deep learning is a path to riches and fame. Due to misleading reports of million dollar salaries for new PhDs and so on. It's a degree with a hi
859.
by currymj 8y ago
The brief portrayal of "jacking in" to cyberspace isn't as weird and psychedelic as it ought to be, but it's better (better = more like the descriptions in Neuromancer) than in any other game I've seen. Everything i
860.
by currymj 8y ago
it is weird that everyone uses this relatively old EMD solver. in this case the WGAN doesn’t actually compute the discrete EMD like that. instead it uses some constraints in the optimization process (gradient clipping), which it can be argu
861.
by currymj 8y ago
do you object to the point of a boycott for even interoperating with CUDA? because PyTorch is still quite pleasant to use in CPU mode. it should also have support for AMD chips quite soon (and AMD's CUDA-equivalent, ROCm, at least appe
862.
by currymj 8y ago
i also found it very frustrating. it reminded me of that famous article about the strange ritual practices of the Nacirema people. (which, if you don't feel like reading it, is actually just an account of 20th-century American life, de
863.
by currymj 8y ago
80,000 Hours is a website made by members of a very particular subculture. The “prototypical” PhD they’re imagining is in computer science, probably with a focus on machine learning, probably at a highly-ranked school, by someone who went t
864.
by currymj 8y ago
yeah, if you have a spam detector that outputs a probability, then that's a rating of trustworthiness on a scale of 0 to 1.
865.
by currymj 8y ago
From reading their website, each bot gets its own neural network and its own reward function, so in that sense they are 5 separate agents. When training starts out, the bots solely focus on their own reward functions. This is so they can le
866.
by currymj 8y ago
I think Rust makes a lot of very hard things possible, at the expense of making some easy things more annoying. That's a tradeoff that bothers a lot of people who hear the Rust hype, only to find that it's more annoying to write t
867.
by currymj 8y ago
it’s missing a lot of the finer details that we are aware of, too. a big one is timing — neurons in the real brain can delay firing depending on what signals they receive. there is research in training artificial neural networks with timing
868.
by currymj 8y ago
all these different interpretations suggest that lots of people, myself included, have a fuzzier understanding of the bias-variance tradeoff than we'd care to admit.
869.
by currymj 8y ago
if you have any interest in baseball, https://www.retrosheet.org/ might be of interest. They have play-by-play accounts for every Major League baseball game going back to 1914. this strikes me as a good testbed for personal
870.
by currymj 8y ago
I interpreted “high-bias” as meaning it has an idiosyncratic and specific audience, but “low variance” as meaning with high probability it will be an effective teaching tool for that audience.

More ›