SergeyHack

7 ms·

1.
by SergeyHack 2y ago
For me it feels like Google is adding "buy" keyword to every search. Part of the answer is here - https://www.wheresyoured.at/the-men-who-killed-google/
2.
by SergeyHack 2y ago
The same here. It helps my slow autistic context switching a lot. It's easier for me to recognize the ideas, concepts, keywords it has written than to recall them from memory when my "memory cache" is empty.
3.
by SergeyHack 4y ago
> You absolutely do feel the same feelings consuming porn as you do actually engaging in sex I can't believe someone says this. It's so not true in my experience. These feelings have a lot in common, but they are definitely not
4.
by SergeyHack 4y ago
+1. Here in Ukraine we see the same reasoning in Russian posts written by Kremlin trolls while they wage their information war against us.
5.
by SergeyHack 5y ago
The usual companies may not have ethics as you write, but politically motivated company can easily embed "ethics" in a recommendation engine like that.
6.
by SergeyHack 6y ago
Vipassana meditation to reveal fears and other mental processes that are hidden. Also to repair focus disseminated by the modern culture of constant distractions. Even 15 minute session is enough to feel the potential and become more aware.
7.
by SergeyHack 7y ago
Interesting, could you give more details on that?
8.
Browser activating the front-facing camera: Big Brother or just a bug(medium.com)
2 points by SergeyHack 7y ago|0 comments
9.
by SergeyHack 9y ago
Really great end-to-end write up. Just a few things: in general case it's better not to use MSE after sigmoid due to slow convergence. And "logits" variable is not logits actually, it's probabilities. Logits is what you
10.
by SergeyHack 10y ago
Sounds interesting. Can these tools work on character n-grams as FastText does?
11.
by SergeyHack 10y ago
Not quite about creating synonyms, but in the same area there is Semantic Vectors https://github.com/semanticvectors/semanticvectors . They process Lucene index and create embedded representation of it. Then you can sea
12.
by SergeyHack 11y ago
Imagine live edit of your video conversations, that attaches joyful emotions to any mention of an advertised brand.
13.
by SergeyHack 11y ago
Smarter agriculture. How to make Permaculture commercially viable while keeping it as ecological as possible. The current agriculture is quite bad for soils and puts many chemicals into our food.
14.
by SergeyHack 11y ago
I've heard a number of times that proper water fasting can heal diabetes. P.S. Absolute fasting can be dangerous.
15.
by SergeyHack 11y ago
I like this one "Why Android Will Soon Become Apple’s Most Important Thing I’ve Ever Done."
16.
by SergeyHack 11y ago
The 6th point is very interesting: "Adversarial examples generalize across models trained to perform the same task, even if those models have different architectures and were trained on a different training set."
17.
Deep Learning Adversarial Examples – Clarifying Misconceptions(kdnuggets.com)
1 points by SergeyHack 11y ago|1 comments
18.
by SergeyHack 11y ago
"HOW TO CRAWL MARKETS" section has good tips for general crawling as well.
19.
MIT Invented a Way to Automatically Fix Software Bugs with Borrowed Code(gizmodo.com)
7 points by SergeyHack 11y ago|0 comments
20.
by SergeyHack 11y ago
To really feel the difference you have to work on complex logic. It's easy to be under illusion that you can create everything using a verbose language. At some point the complex system you work on just gets extremely hard to reason ab
21.
by SergeyHack 11y ago
I wonder what new possibilities for total spying this approach enables.
22.
Blind Motion detects car turns, overtakes, etc. using mobile sensor data(github.com)
4 points by SergeyHack 11y ago|0 comments
23.
by SergeyHack 12y ago
You can try to find a dataset that contains the equal number of positive and negative documents (sentences, etc.) and use it as the validation set. I.e. to tune your hyperparameters on it. In the simple case your hyperparameter can be α in
24.
by SergeyHack 12y ago
I like your "2." suggestion more, because the initial sentiment score distribution can be not normal. So there is an option to try making it normal by taking logarithm for example and calculating mean, etc. after that.
25.
by SergeyHack 12y ago
I wonder if doodling in this context is no more than a weak alternative to walking or other physical activity.
26.
by SergeyHack 12y ago
https://github.com/sytelus/HackerNewsData
27.
by SergeyHack 12y ago
Have you considered sharing the data as a torrent? The current FileDropper speed is quite low (~200KB/s) P.S. Thanks for the data, you saved a good amount of our time.
28.
by SergeyHack 12y ago
According to an interesting book of a Russian paleontologist Kirill Eskov [1] there were a lot of "evolutionary tries" to create a multicellular life. And as soon as it became "profitable" the regular try conquered the w
29.
by SergeyHack 13y ago
The default example "justin bieber - man + women" was ok, but I have found a better one - "justin bieber - women + man "
30.
by SergeyHack 13y ago
BTW, your motto is not perfect IMHO. "Experts" and "expertise" in "Find Experts. Sell Expertise" is too general. I would try to come up with more concrete examples though I realize it's not an easy task.

More ›