Post #460
213
jQuery is so 2006...
http://joelgrus.com/2016/05/23/fizz-buzz-in-tensorflow/
http://joelgrus.com/2016/05/23/fizz-buzz-in-tensorflow/
Channel Public Channel
LI Linkstream
@linkstream
Various links I find interesting. Mostly hardcore tech :) // by @oleksandr_now. See @notatky for the personal stuff
- Subscribers
- 169
- Photos
- 37
- Videos
- 3
- Links
- 966
Showing posts older than #461 · Back to latest
Older Posts 20 shown
Post #459
224
Post #458
209
License Zero: An interesting take on the Open Source sustainability problem
https://writing.kemitchell.com/2017/09/12/The-License-Zero-Manifesto.html
https://writing.kemitchell.com/2017/09/12/The-License-Zero-Manifesto.html
Post #457
234
Post #456
259
Post #455
202
Live International Space Station telemetry
https://isslive.com/displays/index.html
https://isslive.com/displays/index.html
Post #454
199
every technology sufficiently advanced becomes programmable, intentionally... or not.
https://www.blackhat.com/docs/us-17/thursday/us-17-Tsai-A-New-Era-Of-SSRF-Exploiting-URL-Parser-In-Trending-Programming-Languages.pdf
https://www.blackhat.com/docs/us-17/thursday/us-17-Tsai-A-New-Era-Of-SSRF-Exploiting-URL-Parser-In-Trending-Programming-Languages.pdf
Post #453
237
Soviet Control Rooms
http://blog.presentandcorrect.com/27986-2
http://blog.presentandcorrect.com/27986-2
Post #452
207
Julia seems to be quietly amassing momentum
http://www.stochasticlifestyle.com/comparison-differential-equation-solver-suites-matlab-r-julia-python-c-fortran/
http://www.stochasticlifestyle.com/comparison-differential-equation-solver-suites-matlab-r-julia-python-c-fortran/
Post #451
165
I was surprised (though it's obvious in hindsight) to discover that noise removal always results in information loss and hence can't always help legibility — though under certain conditions it sure helps
RNNoise: Learning Noise Suppression
https://people.xiph.org/~jm/demo/rnnoise/
RNNoise: Learning Noise Suppression
https://people.xiph.org/~jm/demo/rnnoise/
Post #450
164
one more point against large, open offices
https://www.ncbi.nlm.nih.gov/pubmed/21528171
https://www.ncbi.nlm.nih.gov/pubmed/21528171
Post #449
184
Post #447
195
Security is hard. Security is not composable, either.
TLDR: Abusing CPU power management to completely take over ARM TrustZone (and probably Intel SGX too), by inducing physical faults while trusted code is running.
https://www.usenix.org/conference/usenixsecurity17/technical-sessions/presentation/tang
TLDR: Abusing CPU power management to completely take over ARM TrustZone (and probably Intel SGX too), by inducing physical faults while trusted code is running.
https://www.usenix.org/conference/usenixsecurity17/technical-sessions/presentation/tang
Post #446
190
Post #445
204
Post #444
192
see also: http://stateofwebtype.com/
Post #443
180
CSS3 Font-Feature-Settings playground
Kerning, Ligatures, Character Alternates, Stylistic Sets, etc.
http://clagnut.com/sandbox/css3/
Kerning, Ligatures, Character Alternates, Stylistic Sets, etc.
http://clagnut.com/sandbox/css3/
Post #442
203
a useful technical report, showing the amount of work that separates "shiny new ML model" from even a prototype product
A Deep Reinforcement Learning Chatbot
https://arxiv.org/pdf/1709.02349.pdf
A Deep Reinforcement Learning Chatbot
https://arxiv.org/pdf/1709.02349.pdf
Post #441
212
hmmm, it stabilizes faster than I thought
https://reasonml.github.io/community/blog/#messengercom-now-50-converted-to-reason
https://reasonml.github.io/community/blog/#messengercom-now-50-converted-to-reason
Post #440
229
The Mating Rituals of Deep Neural Networks: Learning Compact Feature Representations through Sexual Evolutionary Synthesis
https://arxiv.org/abs/1709.02043
https://arxiv.org/abs/1709.02043