tiarkrompf

5 ms·

1.
by tiarkrompf 6y ago
Great question! Unary operators are really simple to add: you just look for the operator symbol first thing at the right level of precedence. Same idea as "if (peek == '(') ..." for parentheses, but outside the code that
2.
by tiarkrompf 6y ago
I just pushed a fix for Chrome (apparently it doesn't report affected selection ranges for input events). Hope it works now!
3.
by tiarkrompf 6y ago
Thanks - very glad to hear. RSS should be doable
4.
by tiarkrompf 6y ago
Author here - happy to answer questions, as always. Thanks also for feedback on the format of the article. It's a bit of an experiment on how to present dense information effectively. Some more rationale here: https://tiarkr
5.
by tiarkrompf 7y ago
For starters, what if you want to differentiate through code that traverses a collection, for example: val x = ... ys.map(y => y + x) Each loop iteration needs to contribute a gradient update to x, which is defined in an out
6.
by tiarkrompf 7y ago
Hi Noel, can you clarify how this corresponds to reverse mode AD? To me it seems more like a delayed or on-demand version of forward AD.
7.
by tiarkrompf 7y ago
Co-author here - nice to see this discussed and happy to answer questions
8.
by tiarkrompf 8y ago
Thanks! Fixed now.
9.
by tiarkrompf 8y ago
Thanks! Yes, it wasn't obvious at all when we started looking at AD either. Lantern supports a good deal of PyTorch (via Snek, our Python front-end similar to AutoGraph) and can also read ONNX. Full feature parity is not our main goal-
10.
by tiarkrompf 8y ago
It's different -- instead of returning a closure, we take a closure as additional parameter (check paper for details). This means that the call stack stays intact and all intermediate values can be stack-allocated.
11.
by tiarkrompf 8y ago
We also have a more recent and slightly longer draft with additional explanations and GPU training results for ResNet 50 and DeepSpeech2: "Demystifying Differentiable Programming: Shift/Reset the Penultimate Backpropagator"
12.
by tiarkrompf 8y ago
Co-author here. Happy to answer any questions, as usual ...
13.
by tiarkrompf 8y ago
No surprise here, as Martin was my PhD advisor :)
14.
by tiarkrompf 8y ago
Hi all, Author here, great to see our paper on the front page today. Happy to answer any questions! Since Spark was already mentioned in the comments, you might be interesting in the parallel/follow-on work we did applying similar tech
15.
by tiarkrompf 8y ago
Thank you -- made my day!
16.
by tiarkrompf 9y ago
Sounds interesting. About the role of CPS: for specialization purposes, CPS can have some benefits by splitting control flow. For example, if your input is "if (c) a else b" and c is a staged (=symbolic) value, then a CPS interpre
17.
by tiarkrompf 9y ago
I'm one of the authors. Happy to answer questions.
18.
by tiarkrompf 10y ago
It's a nice piece of code. I like to use it to explain how to turn interpreters into compilers through staging/specialization: https://scala-lms.github.io/tutorials/regex.html
19.
108: Make a beat in the browser(martinwecke.de)
6 points by tiarkrompf 10y ago|0 comments
20.
by tiarkrompf 11y ago
The talk was part of a summer school at EPFL: http://vjovanov.github.io/dsldi-summer-school/ Afaik, everything was videotaped. I'm sure the videos will appear soon.
21.
by tiarkrompf 12y ago
"precision" JIT: I'm not sure -- can you tell it what to do? More predictable performance model than HotSpot: absolutely.
22.
by tiarkrompf 12y ago
(author here) Yes, one of our points is that 'register' is much too low-level, because one can't abstract over it. The same holds e.g. for OpenMP directives. In the paper we propose (among other things) to attach such directi
23.
by tiarkrompf 12y ago
Author here. Thought I should share the github link as well: https://github.com/TiarkRompf/lancet
24.
by tiarkrompf 13y ago
Relite author here. I'd love to learn more about your use cases, feel free to get in touch!
25.
by tiarkrompf 14y ago
Here is a collection of all his papers: http://infoscience.epfl.ch/search?ln=en&p=bagwell