anjc

33 ms·

301.
by anjc 8y ago
I see this argument frequently, that it's actually a rare occurrence, and that foreigners are just misinterpreting a reasonable request to speak the local language. Well I've tested this. When you're confronted by owners sayi
302.
by anjc 8y ago
Any non-Japanese person who wants to work there should think long and hard, and give it a trial first. Poor pay, strong work culture, stifling weather and crazy weather events for much of the year...it's no paradise. You'll freque
303.
by anjc 8y ago
1) "Google’s AutoML highlights some of the dangers of having an academic research lab embedded in a for-profit corporation" 2) "We can remove the biggest obstacles to using deep learning ... by making deep learning easier to
304.
by anjc 8y ago
> These aren't examples of just productionizing a PhD thesis His point is that sometimes research bears fruit, and sometimes it doesn't. All universities (in my country) now have metrics for how much research must successfull
305.
by anjc 8y ago
Now that's making me nostalgic
306.
by anjc 8y ago
I don't know why the downvotes. Publishing something takes a lot of directed effort. It's so tough to stop doing (what feels like) the real work, and quickly squeeze out a submission for a conference that has a 10% acceptance rate
307.
by anjc 8y ago
Ah I wanted a GP32 too. The homebrew scene looked so great.
308.
by anjc 8y ago
Agreed on the nostaliga, but what cash grabs? The 32x and Mega CD might have been failures, but they were pretty innovative I think. The Dreamcast was also the 90s :)
309.
by anjc 8y ago
Man I remember this deal on Lik Sang. I was so sorry I didn't pull the trigger. What a great site that was.
310.
by anjc 8y ago
Very good, well done.
311.
by anjc 8y ago
How is it possible for an 18 year old to be so accomplished? Insane.
312.
by anjc 8y ago
> That's probably not a very common occurence - wouldn't a lanyard do a better job there? Nope, little plastic arms
313.
by anjc 8y ago
>“Computers equipped with this technology only take minutes to autonomously analyze a breath sample that previously took hours by a human expert” The AI is impressive, but I'm equally impressed that a human can sniff someone's
314.
by anjc 8y ago
Oh man, those tilt-shift 3d visualisations...drool
315.
by anjc 8y ago
The sentence you quoted was very precisely worded: https://i.imgur.com/4n59IWT.jpg
316.
by anjc 8y ago
Presumably the labels aren't coming from human's visual classification of the images.
317.
by anjc 8y ago
> I see "Letter Recognition Dataset" from UCI ML repository. What's the hell is that? Can you point to any paper published in a respectable conference (NIPS/ICML/CVPR) in the last 5 years that used it? Why are yo
318.
by anjc 8y ago
I don't understand your point? Of course there are a whole range of datasets, this paper chose some of them, and performed evaluations on them. If you want to reproduce their research, you can use the same algorithms on the same datase
319.
by anjc 8y ago
A lot of 'gaming the system' and backscratching goes on in Academia. Relying on citation counts, journal/conference names, and the oversight of 2 or 3 peers (who may be 1st year PhD students) is a veerrrrry dubious heuristic.
320.
by anjc 8y ago
This is clearly a bad way to tell if a paper is legit, seeing as this particular paper seems to have been published a week ago and will therefore probably not have any citations for close to a year. Judge papers based on their methodology,
321.
by anjc 8y ago
I'm surprised that the comments are so critical here. Regarding the dataset criticisms: these are canonical datasets in ML research, I see no problem with using them for evaluations here. Using canonical datasets helps with reproducibi
322.
by anjc 8y ago
The paper wasn't about deep learning
323.
by anjc 8y ago
How many billions have been raised during ICOs now? 10 billion? 20 billion? This has been seamlessly and safely accomplished via blockchain.
324.
by anjc 8y ago
You don't think that people should be rewarded for risking their time and capital on an eventually-successful niche technology, versus people who haven't risked anything, or contributed anything? Bitcoin isn't magic beans, it
325.
by anjc 8y ago
Just try and send a transaction that a bank doesn't expect, like an international transfer, or a Western Union transaction and see what happens to your account. And this is in a modern, wealthy country. You're spot on, this gives
326.
by anjc 8y ago
Absolutely bizarre.
327.
by anjc 8y ago
Longs aren't out to profit?
328.
by anjc 8y ago
It sounds like you're describing something more like loans, than a market. Short sellers increase market liquidity, and help with price discovery. There's no destruction. This is a good thing, especially for a mega-hyped but under
329.
by anjc 8y ago
It's still in the same downward channel
330.
by anjc 8y ago
It was effective but not optimal. Some people were overjoyed with their 10x on Ethereum, while others were getting 100x on tokens. It was also obviously highly non-optimal if you held through January :)

More ›