shakascchen

5 ms·

1.
Ask HN: Why there's no Emacs paid support providers?
3 points by shakascchen 2y ago|6 comments
2.
How to do versioning for an online video with the same URL?
1 points by shakascchen 2y ago|0 comments
3.
Wittgenstein's Language Games and the Rise of Large Language Models(shaka.today)
3 points by shakascchen 2y ago|0 comments
4.
OpenAI's Journey and the Challenges Ahead: Reflections from an AGI Entrepreneur(shaka.today)
2 points by shakascchen 2y ago|0 comments
5.
by shakascchen 4y ago
Got a feedback that "Sensible" sounds strange. Any single word recommendation for "UI component that can have some of the multimodality?" Thanks.
6.
Proposing Model-Feature-Sensible Pattern(shaka.today)
1 points by shakascchen 4y ago|1 comments
7.
by shakascchen 5y ago
No fun doing it for Chinese, especially for traditional Chinese. I had to install software but on Cloud SQL you can't. You have to do it on your instances.
8.
by shakascchen 5y ago
Then apply holism and the math of dynamical system. A quote from [The Geography of Thought]( https://en.wikipedia.org/wiki/The_Geography_of_Thought ). "...we’ve seen that modern Asians, like the ancient Chinese, vie
9.
Capitalist market is a peer-to-peer scoring system. Spirituality makes it better(shaka.today)
1 points by shakascchen 5y ago|0 comments
10.
by shakascchen 5y ago
I'll definitely become a heavy Nyxt user. First, I want to programmatically control what I see on some websites. For example, facebook, linkedIn, or more websites are too distracting, so I'd like to directly fetch some information
11.
by shakascchen 5y ago
It seems that the discriminator tests if the output of the generator obeys the pattern of a language. Isn't it still possible that after training the generator outputs the text wrong for the input? On the other words, the output is und
12.
Ask HN: What's the pain using current AI algorithms?
1 points by shakascchen 5y ago|0 comments
13.
by shakascchen 6y ago
Actually supervised learning is "learning missing data dimension" by parameter tuning via associative learning rules. Gradient descents are a special case of associative learning rules assuming all data points the same importance.
14.
by shakascchen 6y ago
There would be other issues like how much we can change our sleeping hours and circadian. The article mentioned it's crazy to say everybody has to be the same height, but nutrition and exercise can influence how tall a children is goin
15.
by shakascchen 6y ago
I would take its being played as one of the evidences that humans have the tendency to do creative activities.