Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
flor1s
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
21 ms
·
151.
▲
by
flor1s
9y ago
A few years ago I worked on a similar kind of project as part of my bachelors. We were converting SQL for Oracle to SQL for Microsoft SQL Server. One problem is dealing with features which language A supports and language B doesn't. An
152.
▲
by
flor1s
9y ago
Maybe I'm confused, but I always thought front-end development included graphics design. It seems like this guide does not discuss graphics design at all. I don't just mean Photoshop, but also subjects like when/where to use
153.
▲
by
flor1s
9y ago
Regarding the Blum, Hopcroft, Kannan book, this seems to be a more recent version: https://www.cs.cornell.edu/jeh/book.pdf
154.
▲
by
flor1s
9y ago
Sounds like an HBO (higher professional education) in The Netherlands. Having done a bachelor at an HBO and a masters at a research university (and now doing a PhD at a university in Japan) I can say Computer Science at an HBO is excellent
155.
▲
by
flor1s
9y ago
Ah yes, I've had the same issue with NAO. One of the best ways to grab an object (water bottle or box) seems to be to do a "hug grasp".
156.
▲
by
flor1s
9y ago
It seems like you are right, their business model had me fooled! The robot costs around 1,200,000 yen (roughly 10,900 dollars) for both the robot and the three year service contract.
157.
▲
by
flor1s
9y ago
I've been talking with some people in my lab about that, it seems like a really interesting competition. I'm really curious to see how Pepper will be performing!
158.
▲
by
flor1s
9y ago
I'm not sure if the price in the article is wrong. NAO costs between 5k-15k, but Pepper uses much cheaper hardware and is sold at a loss so it might be possible to buy it for 1.6k. Like you mentioned, Boston Dynamics and Aldebaran have
159.
▲
by
flor1s
9y ago
I guess the jobs I am applying for are rather competitive. I study in Japan and I am applying for jobs here which are open for those who do not speak business level Japanese, and there aren't many jobs like that. I've build pretty
160.
▲
by
flor1s
9y ago
From that link: 'Rubin also went to SoftBank Group Corp. and negotiated a deal to include Essential in the firm's nearly $100 billion technology fund, but the investment was nixed after Apple joined the fund, according to a report
161.
▲
by
flor1s
9y ago
I did my masters in Computer Science (graduated cum laude) and am currently in the third year of my PhD. I applied for a few companies but I am having trouble getting through the coding tests, which are sometimes very practically oriented (
162.
▲
by
flor1s
9y ago
I'm a bit saddened by the sneer at Aldebaran in the article: '“SoftBank may not have struggled as much if they bought a better robotics company” instead of Aldebaran, Takahashi said.' To be a bit hyperbolic: Aldebaran (rename
163.
▲
by
flor1s
9y ago
In Computer Science many innovations are bio-inspired, such as various robots (Gecko robot from Stanford: http://news.stanford.edu/news/2010/august/gecko-082410.html , Harvard's octobot: http://
164.
▲
by
flor1s
9y ago
We used it in my algorithms class in uni, I found it a bit heavy on the math side of things. For me a more code oriented book such as Skiena's The Algorithm Design Manual works better.
165.
▲
by
flor1s
9y ago
I've seen a research lab in Tokyo use Max MSP for controlling the motion of a humanoid robot (using pneumatic actuators). I believe that whenever there is some flow involved it is easier to model it graphically than textually.
166.
▲
by
flor1s
9y ago
"This video contains content from Sony Music Entertainment (Japan) Inc., who has blocked it in your country on copyright grounds. Sorry about that." I live in Japan...
167.
▲
by
flor1s
9y ago
I'm sure this is already being done at a massive scale. It seems like disabling all connectivity when you are not using it would be good for privacy.
168.
▲
by
flor1s
9y ago
Reading the article it also seems the author has a rather sensationalist approach to interpreting research. Some of the studies he refers to do not seem to make the claims that he extracts from them, nor have a sample which resembles the ge
169.
▲
by
flor1s
9y ago
Reminds me of a website focused on failed Kickstarter projects: http://kickended.com
170.
▲
by
flor1s
9y ago
The goal of the simulations can be for example to render some particle system or create some procedurally generated content for use in a game. Many of their tutorials are focused on including their technology in game engines such as Unity.
171.
▲
by
flor1s
9y ago
I guess being self compassionate when you are doing great does not bring a big benefit, but neither does it hurt. However as the article mentions, it's hard to have self esteem when you are not doing that great. As a PhD student I have
172.
▲
by
flor1s
9y ago
I loved Ingress! Of course after spinning off Niantic (the team behind Ingress), that spinoff went on to develop the highly successful Pokemon Go. Pokemon Go contains many of the gameplay elements of Ingress, but the packaging is a lot bett
173.
▲
Essence of Calculus, chapter 2
(youtube.com)
2 points
by
flor1s
9y ago
|
0 comments
174.
▲
by
flor1s
9y ago
That's so beautiful that I have to admit my eyes started to tear a little bit. It's lovely to live in a world in which wonderful people are empowered to touch people all over the world in such a powerful and profound way.
175.
▲
by
flor1s
10y ago
It's funny because in the lecture itself he discussed how the idea of neural networks was once on the chopping board of the course, but they decided to keep it in just to avoid people reinventing the wheel. It's amazing how a lot
176.
▲
by
flor1s
10y ago
I found Malcolm Gladwell amazing to listen to while running: motivating, exciting and interesting.
177.
▲
by
flor1s
10y ago
What is too much though? Backpropagation uses derivatives, some filters in Computer Vision use multivariate calculus. If you want to have a thorough understanding then calculus is necessary. That said, Andrew Ng was quite good at avoiding c
178.
▲
by
flor1s
10y ago
The course is also quite easy to follow without buying the book. I love the exercises in which you are programming an intelligent agent to move through a maze. It reminded me of how we learned programming in university using Karel The Robot
179.
▲
by
flor1s
10y ago
I think there is a pdf draft of the book legally available from http://ai.stanford.edu/~nilsson/QAI/qai.pdf
180.
▲
by
flor1s
10y ago
I think to be successful at machine learning you also need a good understanding of calculus, besides probability and linear algebra.
More ›