doubtfuluser

13 ms·

121.
by doubtfuluser 5y ago
The main point about most causality questions is to answer the question “what if X did not happen / did happen”. Normally you can’t get that information since you are looking at past events where X happened and you can’t change that. I
122.
by doubtfuluser 5y ago
Maybe a future direction would be to train new models to identify plagiarism by training on this information. Use „non matching backtranslations for training classifiers. It’s again the typical cat and mouse game I guess
123.
by doubtfuluser 5y ago
Like stated below, I’m seriously worried about Adversarial examples that fool the network into A) hiding bad pictures being not recognized B) triggering false positives on harmless pictures to discredit people. Opening up the network would
124.
by doubtfuluser 5y ago
Most of the people see the adversary in Apple (or governments), I think there is something else: What about Adversarial Attacks. Let’s assume someone is going to spread regular memes modified as Adversarial examples to generate the same neu
125.
by doubtfuluser 5y ago
Many comments here actually state that RL is really data hungry. That is true, and indeed one of the difficulties to overcome. However there is a lot of work in the field with making use of transfer learning / meta learning to tackle t
126.
by doubtfuluser 5y ago
… in mice
127.
by doubtfuluser 5y ago
I know too little about biology, but given the sheer number of species in the world, does this one single instance really prove this? How likely is it to see this between any two (unrelated) species in the world?
128.
by doubtfuluser 5y ago
While I appreciate biotech articles being on hacker news, this one is nothing I would expect on the frontage. Three of the authors have interest in the company that makes HD-map. This in combination with being linked as a preprint without p
129.
by doubtfuluser 5y ago
I once got nice shots from Mt Tam. You can hardly see it, but with an ok camera doing long exposure you can get some nice pictures - even being so close to SF
130.
by doubtfuluser 6y ago
For my understanding: crisper works by cutting the dna at specific points surrounding the relevant part and then exchanging, right? How likely is it, that in the whole DNA string there are other instances this “surrounding” matches and woul
131.
by doubtfuluser 6y ago
Over that long period of time, isn’t there the effect of star movement at least a tiny bit visible?
132.
Facebook considers Facial Recognition for smartglasses(buzzfeednews.com)
1 points by doubtfuluser 6y ago|0 comments
133.
Facebook Portal adds support for Zoom(engadget.com)
1 points by doubtfuluser 6y ago|0 comments
134.
by doubtfuluser 6y ago
So he suggests something like the “Hambach Declaration on AI”? When it comes to regulation we are fast in the EU ;) https://ec.europa.eu/futurium/en/institutional-matters/decla...
135.
by doubtfuluser 6y ago
And I’m reminded of “oil imperium”... played that for hours - when I was young
136.
by doubtfuluser 7y ago
i Think what they misunderstood is really their customerbase. Mid level audiophiles who want comfortable music with good sound. I don’t care if I can talk with my speaker and order pizza or not. And my rough guess is, that just looking at t
137.
by doubtfuluser 7y ago
So will it be enough to disable automatic updates to prevent Sonos from cripple my system?
138.
Sonos stops support for older devices(cnet.com)
3 points by doubtfuluser 7y ago|2 comments
139.
by doubtfuluser 7y ago
Pretty cool work. Can you give any insights on the silver you used - especially the OSS ones. I had mixed (pun intended) experiences and would love to hear other opinions
140.
by doubtfuluser 8y ago
Ok, so I now have the following question: what if I’m evil (that kind of evil person none can imagine would exist on Facebook), and I impersonate for example someone else - like my ex for example. Will I get notified about every photo they