Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
aqsalose
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
20 ms
·
121.
▲
by
aqsalose
9y ago
>It's up to you to not overdo the designs and the animations That's why I like LaTeX Beamer (at least as long as I don't need anything else in my presentation other than bullet point text and some math): in LaTeX, writing
122.
▲
by
aqsalose
9y ago
Title and the linked story don't match?
123.
▲
by
aqsalose
9y ago
> What software development skills will be most useful in a post-apocalyptic world? At the moment of writing this, North Korea alone is not able to bring about the kind of total apocalypse people are thinking when they talk about "p
124.
▲
by
aqsalose
9y ago
What T-A said. Moreover, Finland as a country exists because Czar Alexander I conquered some eastern provinces of Swedish Realm (as a weird fluke of Napoleonic wars), and then decided that they would be useful as a buffer zone for defense o
125.
▲
by
aqsalose
9y ago
Does the article disagree with any of that? >Slezkine is by no means the first person to draw the analogy between the Bolsheviks and sectarians (Lenin himself is reported to have taken an interest in the Münster Anabaptists and Cromwell’
126.
▲
by
aqsalose
9y ago
I'm going to propose as a conjecture that (maritime) radio weather reports are universally soothing and nostalgic, were they written by Met Office and broadcast by BBC or not. I don't know specifics of the UK culture, so maybe the
127.
▲
by
aqsalose
9y ago
By custom, Greek letters usually mean different things than letters of Latin alphabet. And also it's customary that certain letters in Latin alphabet are used for specific things. For example, Latin letter 'x' usually means a
128.
▲
by
aqsalose
9y ago
As I said when I spotted this on Twitter, I think highlighting multiple things at the same time is a bit confusing. Looking at the equations, I, for one, can't easily spot any connections between the colorized parts of the equation and
129.
▲
by
aqsalose
9y ago
>They smell bad, are unattractive and grouchy. >I certainly won't miss riding the BART with sketchy looking characters at 12am. I wonder if there is difference in the amount of, erm, sketchiness of your sketchy looking characters
130.
▲
50 Years of Data Science
(tandfonline.com)
1 points
by
aqsalose
9y ago
|
0 comments
131.
▲
by
aqsalose
9y ago
>Another example from the annals of machine learning history: time and time again when there is a breakthrough in architectures, it's usually followed a few years later by a simplification of the architecture. For example we started
132.
▲
by
aqsalose
9y ago
See also further discussion on Andrew Gelman's blog: http://andrewgelman.com/2017/11/28/five-ways-fix-statistics/
133.
▲
Ways to fix statistics
(nature.com)
134 points
by
aqsalose
9y ago
|
52 comments
134.
▲
by
aqsalose
9y ago
The Turing award is one of those awards many people in the field know about; each research field has their own nobody else has heard about. Nobels are the award that everybody knows about.
135.
▲
by
aqsalose
9y ago
> Apple did not break a law, so law enforcement cannot hold them accountable. I thought the unstated assumption was that the law should be changed, which is certainly within both the letter and spirit of rule of law.
136.
▲
When man meets metal: rise of the transhumans
(theguardian.com)
5 points
by
aqsalose
9y ago
|
2 comments
137.
▲
by
aqsalose
9y ago
Having read some very introductory stuff about human visual system, I'd be willing to guess that a partial reason is that the human vision works on the input level quite differently than "observe NxM colored pixels in grid".
138.
▲
by
aqsalose
9y ago
Article cites one hard number: >Global figures are hard to come by, but only three or four in every hundred PhD students in the United Kingdom will land a permanent staff position at a university. It’s only a little better in the United
139.
▲
by
aqsalose
9y ago
Is there a sane solution to following local timelines of several instances? The local timeline on my instance is quite dead, the federated timeline has so much variance it's nearly useless.
140.
▲
by
aqsalose
9y ago
Sure, it wouldn't do any harm to move to relevant and well-sourced examples, or construct such demonstrations specifically for the class. This HN discussion alone has gathered several instances of classifiers overfitting or fitting to
141.
▲
by
aqsalose
9y ago
I have heard this story multiple times. My impression is that often person who tells the story treats the fact that the mistrained model was NN-based as a minor detail (or the kind of juicy but ultimately insignificant detail that make the
142.
▲
by
aqsalose
9y ago
This is the PhD thesis of Durk Kingma [1]. Reading this right now; appears to be a well-written introduction to stuff variational autoencoders [2] and ADAM [3]. [1] http://dpkingma.com/ [2] http://arxiv.org/
143.
▲
Variational Inference and Deep Learning: A New Synthesis [pdf]
(dropbox.com)
1 points
by
aqsalose
9y ago
|
0 comments
144.
▲
Alice in Python projectland [Python packaging tutorial]
(veekaybee.github.io)
3 points
by
aqsalose
9y ago
|
0 comments
145.
▲
by
aqsalose
9y ago
Problem is that the useful societal networks as described by e.g. the other comment below [0] are naturally scarce resources. A fraternity (and the associated wider friends-of-friends social network) that consists of dozens or even hundreds
146.
▲
by
aqsalose
9y ago
"All the Linear Algebra You Need". The title is misleading. In general, I like an applications-first approach, but I have hard time liking a tutorial with the title like that and which proceeds to say things like >Neural networ
147.
▲
by
aqsalose
9y ago
I think the main difference is not the media, but the fact that the US government acknowledges that the incidents really happened, and that they are investigating.
148.
▲
by
aqsalose
9y ago
And as the story goes, after reading an obituary that argued exactly that ("reports of my death are greatly exaggerated" kind of situation), Nobel drafted a peculiar testament about how his fortune should be used after his death.
149.
▲
by
aqsalose
9y ago
Semi-related: is there any easy way to follow local timelines of multiple instances with a single client? My current main instance has only handful of active people, so I've been wondering if I should look at somewhere else. However, I
150.
▲
by
aqsalose
9y ago
Previous discussion about Mastodon: https://news.ycombinator.com/item?id=15211074
More ›