Hacker Newsnew | past | comments | ask | show | jobs | submit | knoble's commentslogin

Would you mind sharing any of the resources you are using for re-learning? I've been meaning to do the same.


For probability/statistics you could also use the MIT Course https://www.edx.org/course/introduction-probability-science-...

Same course if you prefer the classroom lectures http://ocw.mit.edu/courses/electrical-engineering-and-comput...

Or if you want more rigor you can go through these notes that cover the same material but in a more formal way (via sigma algebras and measure theory) http://ocw.mit.edu/courses/electrical-engineering-and-comput...


Just saying, but if you want to hop onto the ML bandwagon (for instance), then don't bother going over linear algebra or probabilities first, and instead just learn what you need as you go. For example, the first sections of this book are already devoted to getting you on the right track, and it's somewhat standard to do so. And besides, there's no need in learning what are rotation matrices if you won't use them.


As a counterpoint, if parent is interested in taking ML further, a solid foundation in linear algebra will be huge when more advanced signal processing applications come up.


A couple of friends recommended these:- (Not sure if they are relevant though for deep learning specifically)

1) http://ocw.mit.edu/courses/mathematics/18-06-linear-algebra-... 2) https://www.khanacademy.org/math/linear-algebra/vectors_and_...

If anyone knows anything else (relevant to deep learning) could you please share :)


Introduction To Statistical Learning:

http://www-bcf.usc.edu/~gareth/ISL/

Is an excellent statistical learning reference.


I've been going through this series of video lectures on Youtube:

https://www.youtube.com/playlist?list=PL5102DFDC6790F3D0

for a basic "Stats 101" course.

There's also this archived Coursera course. There aren't any active sections to sign up for, but the videos are still available:

https://class.coursera.org/introstats-001


I'm mostly using Khan Academy at the moment. But I see several people already posted alternatives which is nice to have ;)


* Foundations of Machine Learning

* All of Statistics

* Doing Bayesian Data Analysis

Also the ML specialization on Coursera


Mark Cuban did this when selling broadcast.com to Yahoo. It's a pretty amazing hedge: http://investmentxyz.blogspot.com/2006/05/cubans-collar-anat...


First dates are probably most analogous to phone interviews. It's a testing of the water. If you've met the barrier of entry, you'll probably make it to the next one unless there is no chemistry whatsoever.


I think the main point is that it is dangerous to engage in spoken-word type contexts publicly because there is no delineation between contexts that should be treated as conversational and those that should not. Conversations provide the context to say something dumb, realize it is dumb, and then learn from it without the public shaming and castigation. Tweets are eternal, whether through screens caps or similar preservation techniques.

The most compelling argument for me was about context collapse. Shifting between spoken-word contexts and written-word contexts is at the core of the issue that requires a feature like moments. Spoken-word contexts lack an explicit connection to the events they discuss and require manufacturing a context. I think moments has been great at doing this so far but its existence is a symptom of the bigger problem.


http://www.npr.org/sections/money/2014/02/12/275922026/episo...

http://regressing.deadspin.com/how-much-is-lebron-really-wor...

http://www.sbnation.com/lookit/2013/5/1/4291088/lebron-james...

And many more...

It is generally agreed upon that in a salary capped league, and especially in basketball wherein a single player has a disproportionate impact on the outcome of a game, salaries are depressed in comparison to what teams would pay for his services on an open market. Lebron guarantees a competitive team no matter where he plays and significantly increased revenues and franchise values when leaving for Miami and returning to Cleveland.


I joke sometimes that its the intersection of nominally left/liberal American pundits with sports fans. The political blog Lawyers Guns and Money being a good example of it-people you wouldn't think would have those views, but when it comes to professional sports, all of a sudden self proclaimed socialists love themselves a free market.


It's not a love of the free-market that drives these conversations; it's a hatred of exploitive capitalism. Ultimately, the players are source of all the league's revenue, yet the have to split that money with the owners (who contribute almost nothing to the day-to-day revenue stream).

The players also have a strong union, which negotiates a number of protections for the players, from minimum salaries to safe work conditions, while allowing for higher compensation for key players.

So it is very reasonable for those of us on the left to be against the extreme labor exploitation represented by the salary cap.


Can players own a team instead of the traditional owner model?


Theoretically it's possible. Realistically no. The leagues vote on team ownership transfers, and they are unlikely to allow a sale that upsets their apple cart. Which is why only one team (Green Bay Packers) out of the big three sports is owned by a government body.


I don't think the problem with scaling is as much economic as it is a product of the number of disparate motivations at scale.

The more people you have, the more difficult it becomes to have any sort of general consensus.


This is really helpful. Thanks for sharing!


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: