Machine Learning Street Talk

it is funny how people with liberal tendencies resort to "it is a free market" when it is convenient to the them and if not they say it is rigged for people with power

That was exactly my point. It being a “free market” is why it sucks in the first place — but if it’s going to be a shitty adtech-driven brain-sludgifying monstrosity, the very least we can do is not actively promote the worst things people put into it.

The same will cry for net neutrality and government provided internet. Go start your own ISP if you don't like Comcast (or whoever).

Two competing camps trying to fork the nation because they both believe the guys upstream keep making shitty commits, year after year. Unless those 9 reviewers properly do their job, it's only a matter of time before the whole project is deprecated.

When should I use XMPP and and RPC ?

I really liked the approximation algorithms from a french university

Investor visas are not terrible. EB5 is great provided you have 500k to put in a single investment.

Well that is much more than I have, I am planning to go with 100k, is there any other program or something? I pretend to relocate. Thanks!

The 500K minimum is to get a green card. All you really need is a E-1 visa, which Argentina seems to qualify for. That visa has no minimum investement, so with $100K, you should easily be able to open a US company, and start a legitimate business. And get your visa.

In many cities and countries the taxi industries have successfully ‘captured’ the regulatory authority, and now the regulations mainly exist to restrict competition.

Because they're not keen on people whose business model is: "having laws is a cost we don't need to pay".

People still listen to Naomi Wolf? I was really wanting to like the "Shock Doctrine" but being from South America, I found the history she portrayed of SA to be totally inaccurate.

That's Naomi Klein.

There have always been "market-makers" why do you care if they are algorithmic or human ?

why is the volatility a factor ? My intuition say that you with high volatility is a lot easier to pick the wrong equities. Is that correct ?

There are two ways to compare returns. Higher returns at the same risk or the same returns at lower risk. Either is better.

volatility is the de facto measure of risk for a portfolio/trading strategy (there are more measures but vol is common). People tend to measure performance on a risk-adjusted basis (for example the Sharpe or Information Ratio) because in an ideal world an investor could theoretically lever/delever a portfolio to a level of risk they are comfortable with. All things being equal a higher Sharpe ratio would give you more returns for the same amount of risk as a portfolio with a lower Sharpe ratio.

Can you point to resources/books where I can read more about design as is taught in CS degrees? I'm curious on their problem solving strategies compared to mine.