anishathalye

11 ms·

61.
by anishathalye 7y ago
Basalt as it exists today already integrates pretty well with LaTeX -- I have papers written in LaTeX that have Basalt figures. The figures are rendered into SVGs, converted to EPS using Inkscape's CLI, and placed in the paper with \in
62.
by anishathalye 7y ago
Thanks :) I'm currently using SMT for a number of reasons. One issue is that I've found the constraint is often non-convex, and this sometimes causes issues. And yeah, as you guessed, the initialization is also a hard part. With s
63.
by anishathalye 7y ago
Hi HN! I've been frustrated with standard GUI-based design tools, because they don't match the way I think: like a programmer, in terms of relationships and abstractions. (Does this resonate with the community here?) So I've
64.
Experiments in Constraint-Based Graphic Design(anishathalye.com)
211 points by anishathalye 7y ago|42 comments
65.
by anishathalye 7y ago
I think one reason some people prefer symbolic links over directly managing $HOME with git is that there's less chance that things go wrong due to a bad accidental git command. For example, `git clean` (with the appropriate arguments,
66.
by anishathalye 7y ago
Some extra info/links that might be interesting to people: 6.S081/6.828, MIT's Operating Systems class, was revamped for this semester. The professors ported xv6, one of the teaching operating systems used for the course, to
67.
by anishathalye 8y ago
For most of the videos, we do have screen recordings (e.g. see https://www.youtube.com/watch?v=VW2jn9Okjhw ). Sadly, we're missing screen recordings for a couple videos (such as the shell video, where we had some issues
68.
MIT Hacker Tools: a lecture series on programmer tools(hacker-tools.github.io)
1028 points by anishathalye 8y ago|79 comments
69.
by anishathalye 8y ago
I made one of these GPS -> WWVB simulators a couple years ago. There's a short write-up here, in case anyone is interested in reading about it: https://www.anishathalye.com/2016/12/26/micro-wwvb/
70.
by anishathalye 8y ago
That's a great idea -- I've submitted it to Overleaf.
71.
by anishathalye 8y ago
Hi HN, author here. I wanted to give a little bit of background on the motivation behind this post. For a while, I've been making academic posters using PowerPoint, Keynote, or Adobe Illustrator, and while it's possible to get a h
72.
Show HN: Gemini, A Modern LaTeX Poster Theme(anishathalye.com)
6 points by anishathalye 8y ago|3 comments
73.
A Brief Introduction to Adversarial Examples(people.csail.mit.edu)
4 points by anishathalye 8y ago|0 comments
74.
by anishathalye 8y ago
Hi HN! I'm one of the authors of this resource. Here's a little bit of background on why we created this resource: Adversarial machine learning is a relatively new but rapidly developing field. It's easy to see why people are
75.
Black-Box Adversarial Attacks with Limited Queries and Information(labsix.org)
1 points by anishathalye 8y ago|0 comments
76.
by anishathalye 9y ago
It's public now: https://github.com/bijection/sistine It's currently lacking detailed instructions and stuff, but you can expect to see more in the coming week.
77.
by anishathalye 9y ago
We just made the source code public: https://github.com/bijection/sistine We're planning on adding more instructions to the GitHub repo over the coming week, to make it easier for people to replicate our hardware
78.
by anishathalye 9y ago
Hi HN! This blog post is about Project Sistine, a hack that I worked on with @antimatter15, @bijection, and @logane. We turned a MacBook into a touchscreen using only $1 of hardware (a small mirror, some pieces of a rigid paper plate, and a
79.
Turning a MacBook into a Touchscreen with $1 of Hardware(anishathalye.com)
239 points by anishathalye 9y ago|16 comments
80.
by anishathalye 9y ago
Hi HN! I’m one of the researchers who worked on this. Both Nicholas Carlini (a co-author on this paper) and I have done a bunch of work on machine learning security (specifically, adversarial examples), and we’re happy to answer any questio
81.
Breaking 7 / 8 of the ICLR 2018 adversarial example defenses(github.com)
2 points by anishathalye 9y ago|1 comments
82.
Partial Information Attacks on Real-World AI(labsix.org)
4 points by anishathalye 9y ago|0 comments
83.
by anishathalye 9y ago
Goodfellow et al. demonstrate that adversarial misclassification works in the physical world: https://arxiv.org/abs/1607.02533 -- a printed image is consistently misclassified (demo here: https://www.youtube
84.
by anishathalye 9y ago
It's probably referencing Szegedy et al. 2013 ( https://arxiv.org/abs/1312.6199 ).
85.
by anishathalye 9y ago
Yeah, totally agree that it could be useful to the community -- open-sourcing it under a permissive license is on our todo list :) It's a reasonable amount of effort to disentangle it from the rest of our code and build a nice API, so
86.
by anishathalye 9y ago
Yeah, we implemented a differentiable renderer.
87.
by anishathalye 9y ago
Haven't had time yet. Also, I haven't decided whether I want to extend it and try to turn it into a commercial side project.
88.
by anishathalye 9y ago
We got all our prints through zverse.com, and we used full color CYMK printing. We didn't correct for color artifacts; we just used a wide distribution with EOT (we just modeled color inaccuracy rather than trying to make a color map o
89.
by anishathalye 9y ago
Thanks! The unperturbed baseball also looks like that -- the model we found is of a dirty baseball :P We probably should have included a photo of that in our blog post, but you can see our paper if you're interested in seeing what the
90.
by anishathalye 9y ago
We only modify the texture; we retain the original shape of the object. If you look closely and compare the unperturbed original with the rifle turtle, you can spot some differences.

More ›