Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
bmc7505
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
36 ms
·
361.
▲
by
bmc7505
8y ago
A Mathematical Theory of Codesign: https://arxiv.org/pdf/1512.08055.pdf
362.
▲
The Principles Behind Differentiable Programming
(youtube.com)
3 points
by
bmc7505
8y ago
|
0 comments
363.
▲
by
bmc7505
8y ago
Not sure what you mean. The paper above claims 1000x speedups for computing second-order derivatives. Have not tested their claims, but was speculating that such an improvement, if true, would make computing hessians for small networks fesi
364.
▲
by
bmc7505
8y ago
Just out of curiosity, but why doesn't Raspberry Pi 3B support simultaneous broadcasting and receiving of Wi-Fi? I've seen multiple apps that swap between these modes (e.g. using broadcast mode to set up a captive portal) and then
365.
▲
by
bmc7505
8y ago
If their chief scientists start disappearing and their colliders suddenly malfunction, I’m going to be pretty disappointed. Trisolarans had better watch out, our Space Fleet is waiting for you this time.
366.
▲
by
bmc7505
8y ago
Would be interesting to explore whether evaluating ∇²f(x) for higher-order SGD methods directly is now feasible for smaller DNNs and whether this leads to faster convergence during training. Most methods like Newton or Gauss-newton were tho
367.
▲
by
bmc7505
8y ago
Not 3X. Three orders of magnitude faster. If true, this is indeed, an exciting result. Maybe as exciting as Strassen’s algorithm was for matrix multiplication - this enables a bunch of powerful optimization methods that were previously t
368.
▲
by
bmc7505
8y ago
If you don't need 3D, 2D LIDAR is pretty cheap right now: https://www.adafruit.com/product/4010
369.
▲
Deep Learning Type Inference [pdf]
(vhellendoorn.github.io)
2 points
by
bmc7505
8y ago
|
0 comments
370.
▲
by
bmc7505
8y ago
"It doesn't matter how beautiful your theory is, it doesn't matter how smart you are. If it doesn't agree with experiment, it's wrong. In that simple statement is the key to science." -Feynman
371.
▲
by
bmc7505
8y ago
AD is important to understand for ML practitioners in the same way as compilers are important to understand for programmers. You can get away without knowing all the details, but it helps to understand where your gradients come from. Howeve
372.
▲
by
bmc7505
8y ago
This paper won a best paper award at ICFP 2018. Conal Elliott recently gave a longer talk about it at Microsoft Research: https://www.youtube.com/watch?v=ne99laPUxN4
373.
▲
The Simple Essence of Automatic Differentiation [pdf]
(conal.net)
167 points
by
bmc7505
8y ago
|
16 comments
374.
▲
The AI Cold War That Could Doom Us All
(wired.com)
11 points
by
bmc7505
8y ago
|
0 comments
375.
▲
by
bmc7505
8y ago
I still think Total Annhilation had one of the best soundtracks of any video game ever recorded, including The Elder Scrolls. It was Jeremy Soule’s magnum opus. https://youtu.be/o2jS4ZaVOWA https://youtu.be/
376.
▲
by
bmc7505
8y ago
The idea is that in the not-too-distant-future, differentiable programming will become much more mainstream. c.f. https://medium.com/@karpathy/software-2-0-a64152b37c35
377.
▲
by
bmc7505
8y ago
Julia has some form of typing checking, but it is not statically typed. https://docs.julialang.org/en/v1/manual/types/index.html
378.
▲
First-Class Automatic Differentiation in Swift
(gist.github.com)
111 points
by
bmc7505
8y ago
|
43 comments
379.
▲
by
bmc7505
8y ago
Nice parallel with Iverson's 1979 Turing Award Lecture, "Notation as a Tool for Thought": http://www.eecg.toronto.edu/~jzhu/csc326/readings/iverson.pd...
380.
▲
by
bmc7505
8y ago
Awesome! I look forward to trying out the speech-to-intent and speech-to-text products. Will these also be Apache 2.0 licensed?
381.
▲
by
bmc7505
8y ago
It would be nice to support a grammar, a la CMUSphinx: https://cmusphinx.github.io/wiki/tutoriallm/#grammars
382.
▲
by
bmc7505
8y ago
Rachael Tatman at Kaggle does a livestream on Twitch every Friday: https://www.twitch.tv/rctatman
383.
▲
by
bmc7505
8y ago
What is the app?
384.
▲
by
bmc7505
8y ago
Qin Shi Huang lifted the sword to the sky, and shouted: “Computer Formation!” Four giant bronze cauldrons at the corners of the platform came to life simultaneously with roaring flames. A group of soldiers standing on the sloping side of t
385.
▲
by
bmc7505
8y ago
Better make sure they don't broadcast anything towards the sun...
386.
▲
by
bmc7505
8y ago
The topmost of notches. Very reproducible and user-friendly. https://devblogs.nvidia.com/docker-compatibility-singularity...
387.
▲
by
bmc7505
8y ago
If you're frustrated with Docker's slow pace of development are are looking for a good alternative, have a look at Singularity containers. They're interoperable with Docker Hub and OCI compatible images, and offer a much bett
388.
▲
by
bmc7505
8y ago
If you're already familiar with iOS, you might enjoy using Kotlin Native. KotlinConf has a demo app for iOS written in Kotlin Native: https://github.com/JetBrains/kotlinconf-app
389.
▲
by
bmc7505
8y ago
Lots of things are self-replicating that we would not consider "living". https://en.wikipedia.org/wiki/Self-replication
390.
▲
How China censors the net: by making sure there’s too much information
(theguardian.com)
4 points
by
bmc7505
8y ago
|
0 comments
More ›