The trees now have some movement! It's gotta get breezy on an isolated island like that... 💨🌳🍃🌅
Bridging discrete optimization with deep learning. Maybe highly optimized (nondifferentiable) solvers for combinatorial problems will become great again 😍.
Differentiation of Blackbox Combinatorial Solvers https://arxiv.org/abs/1912.02175 https://twitter.com/MichalRolinek/status/1203002311292928000/photo/1
Trying to verify the property as stated in textbooks, we have first independently discovered Bloom's mistake and then, in a hard way, also learned that the follow-up correction also had a bug. Having a fully axiom-free proof in Coq we finally know we got it right. 4/6
2/ The sentence “The realloc function returns a pointer to the new object (which may have the same value as a pointer to the old object)” is unfortunate because THERE IS ABSOLUTELY NOTHING YOU CAN DO TO TAKE ADVANTAGE OF THIS.
Yeah, sex is cool but have you ever put the last Qed into your 1800 line proof to see the entire thing light up green with not a single 'admit' line
wishing for visualizer plugins for the proof state, just give me the terms, I'll generate an SVG and the UI can then show it
give it the domain specific representation it needs (in my case that'd be a timeline, xe-xl, oe-l, xe-l being time spans on there)
Excuse me, what?
we interrupt regular super mario kart programming to offer you: a Tennyson solo
I am happy to announce that Debug Entry Values (DW_OP_entry_values) feature is implemented in LLVM! :) It has been enabled by default for x86, arm and aarch64 targets. This should improve debugging user experience when debugging optimized code. #llvm @email@example.com
unimportant compiler dev (@rrika9 on twitter)
Server run by the main developers of the project It is not focused on any particular niche interest - everyone is welcome as long as you follow our code of conduct!