jndsn402

What's involved in building a site like this? I am subscribed to an industry forum in the beverage space where people buy and sell equipment using forum posts - t's really inefficient and I often think they could use a simple marketplace site.

I'm an actuary - from your site bio it appears you may be as well. As you know, P/C insurance carriers have been using machine learning for quite some time. What is it that you are doing differently?

Look into writing apps for Shopify, many are quite simple and charge a fee.

The models as described in the article do seem to have glaring flaws. However, the article did not show that anyone has been killed based on the models alone.

With respect to adversarial examples for random forests, that link states:

   Myth: Deep learning is more vulnerable to adversarial examples than other kind of machine learning.

   Fact: So far we have been able to generate adversarial examples for every model we have tested, including simple traditional machine learning models like nearest neighbor. Deep learning with adversarial training is the most resistant technique we have studied so far.

I don't disagree with the math, but wouldn't it be the case that it's easier to get a biased sample from 1000/100M people than 1000/1M people?

For instance, maybe your sampling ignored a particular region of the country or socioeconomic strata etc.

Best of luck to you, I am not in the industry and cannot recommend from that perspective.

As a consumer though, augmented reality ( eg Hololens and Magic Leap) seems like it is about to become very big, and probably utilizes machine vision.

Can you recommend some resources for learning about web scraping?

Wait But Why on a similar theme: http://waitbutwhy.com/2015/12/the-tail-end.html

HPMOR was fantastic - very well done re-imagining of Harry Potter with Harry as a rationalist, challenging long-held beliefs about magic. Plot, pacing, etc. were great. Really influenced how I think about my own beliefs as well.

Yes.

I worked with a large company preparing for Solvency II. My impression was that it would generally increase regulatory capital requirements, not decrease.

Actuary here, though I don't work with these types of policies.

Your description of how insurance makes money is the standard understanding. However, I would think that most products do have some underwriting gain built in as well.

It can get pretty complex as you need to set up a liability (reserve) to match the assets (premium) that come in. Interest causes both the asset and liability to increase. As benefits are paid out, both the asset and liability decrease. If the liability decreases more than the asset, you make money. Or if you have a substantial surplus (asset - liability), the interest on that is pure profit as well.

Actuary here. I don't have the math at my fingertips, but to confidently estimate the risk of any occurrence you need a significant sample of the event, not just a significant amount of exposure.

So for instance, if you are estimating the mortality rate of a 25 year old and the rate for an 85 year old you would be equally confident in each estimate if your data sample contained an equal amount of deaths for each, not an equal amount of life-years.

So while we have many days of exposure to terrorism risk, we do not have many occurrences and therefore do not have a confident estimate of the rate of major terror attacks.

Just because a smart gun won't solve the problem of gang violence doesn't mean they should not exist.

I don't know personally, but presumably companies designing these guns have thought of the same concerns and made the rings to be durable and sizable.

Re battery dying - if I was designing a smart gun, I would have it default to a regular, usable-by-anyone gun if the power dies. That way you don't have to worry about mastermind criminals with EMP devices either.

What specific case are you afraid of? You are outside in the wintertime with a gun concealed on your person, and a would-be assailant comes along? Then get the kind that uses a ring, not a fingerprint.

And when you lose the ring, get a new one immediately instead of waiting for the would-be assailant to come along.

They are very powerful in that almost no politician is willing to act against their interests. Please correct me if that impression is wrong.

Don't people have guns for self defense? What if someone invades your home in the middle of the night - do you want to have to start fiddling around with your gun vault?

if I'm wearing gloves

Are you currently holding the gun? If not, take off the gloves and then pick up the gun.

if I lose the ring or it comes off

Same thing that happens if you lose your wallet or car keys. You go find it immediately or get a replacement quickly.

girlfriend needs to grab my handgun

Hopefully they can make smart guns that can have two authorized users.

so many ways for this to fail

Regular handguns can also fail - anyone can pick them up and shoot anyone else. Some people consider that a bug, not a feature.

As a parent of small children, if I were ever to buy a gun I would want to have a smart gun.

Seems to me the real obstacle is the NJ law that would restrict regular gun sales once smart guns are sold, which should be revised ASAP as it gives the NRA a semi-legitimate reason to oppose the sale of smart guns.

Did John Draper pioneer the field of maritime breakfast?

Can you define what you mean by an 'actual element of reality'?

I have read Flash Boys but not the books written in response.

How do the other viewpoints explain why the IEX exchange was created and designed specifically to stop HFT? That is the central narrative of the book, and in Lewis's telling it makes complete sense as a way to save the common man from the evil HFT traders. But if Lewis got it all wrong and HFT is perfectly fine, why would a completely new exchange be set up to stop it?

Freelance actuarial consulting - email me at the address in my profile if interested.

In the insurance space I have focused on disability insurance, annuities, financial reporting, and general profit/loss modeling.

As far as modeling is concerned I have built extensive models using combinations of Excel/Access/VBA. Not sure what you mean by data tables - if you are referring to something like actuarial mortality tables, yes I can create those from scratch based on population data.

"...and it makes boring reading."

I've always felt that this was wishful thinking on the part of the mods. Discussions about downvotes (or completely legitimate questions as to why something was downvoted) are no more boring than hundreds of other topics discussed here every day.

If you want to get an understanding of basic SQL without all that overhead, you can try Microsoft Access.

I just mean that married people have experience partnering closely with another person to achieve shared goals, and will have hopefully developed some strategies to do that effectively.

Agreed, much of the startup advice from YC and other outlets is geared toward 20-somethings, and thus may come across as simplistic to people who have:

a) worked in a professional setting (e.g. a large corporation) b) managed other people c) been married d) been alive since before 1980

Thanks for this detailed write-up. I guess I struggle more with the particle aspect than the wave aspect. Meaning, if light were exclusively a wave, then the half-silvered mirror doesn't 'randomly' either reflect or pass through, it just splits the beam in two. And all the events described above would make sense - the beam splits at each half-silvered mirror, so if the block is in place you get hits at B, and if not, you don't. (Correct?)

Except that we are somehow convinced that we are sending single, indivisible photons. Why is that?

And in general - I assume these experiments have actually been done, i.e. at some point long ago someone tried this and was surprised to see that the interference happened with individual photons. Is there a write up of one of these experiments somewhere?