solve

Edit:

You guys are all wrong. There is no blanket automatic acceptance of contracts if you ignore them in the UK.

There IS automatic acceptance of updates to previously signed contracts, ONLY if the original signed contract contains "terms permitting the employer to make changes from time to time" without your permission.

That is completely different than automatic acceptance of all contracts that are left unsigned and ignored.

https://worksmart.org.uk/work-rights/pay-and-contracts/contr...

That sounds insane.

For parents to conduct surveillance? Only if you define "parent" as the "government".

Surprisingly good, both as a broad overview and in the specifics.

Are you implying that YC doesn't require your age in the interview process? Have you ever even filled out the application form? Let's start with that.

Huge difference. CA companies and investors (e.g. YC) specifically require your age to the interviewers. Using YC as an example, many interviewers are happy to publicly boast about using age as primary factor - I can link to to some comments written here on HN.

In some other places, e.g. NYC, this is all generally believed to be highly legally-risky and no respectable employer would ever consider doing it.

Shame on you CA.

That's exactly why I'm surprised that the title of this post suggests that they would be doing this, but they didn't.

Extremely bad title. This is about one startup, not startups.

If they do want to do some properly backtested analysis of the predictive value of various financial indicators versus expected value across many startups though, that would be great.

MOST accelerators will outright reject, if not make the barrier to acceptance vastly higher for any company that went through a "pre-accelerator". Being in a pre-accelerator makes it far harder to get into an accelerator. So pre-accelerator is really a misleading name.

But change the name of these organizations to "VC firms" or "angel groups" - much better description.

You cannot "admit" a prediction of the future as if the prediction of the future were a fact. It can only be predicted.

Tell me more about the creation of the elements that "come from cosmic rays".

Accepting loans no matter what, and paying them no matter what, is actually what will ruin university for our children, and has already been doing so.

Marketplace dynamics are weird like that. They often the OPPOSITE behavior of what you'd expect.

No, it's not for those moral reasons.

- At best, it's for the state to make money in the long term by reducing the brain-drain of great students moving to other states.

- At worst, it's for the university to make more money short-term via price discrimination, by operating under the assumption that people who travel further often tend to have more money to spend than those who stay in state.

Wow, sounds like an actual alignment of incentives.

You're wrong. The US courts weigh nearly always weigh intent unusually highly, compared to the rest of the world.

It wouldn't, because final scores are only evaluated against the "validation" set.

As for turning current contests into an exercise in overfitting the "test" set - we already reached that point long ago. Test vs validation scores often diverge wildly in these contests.

Edit - Replying to arnsholt:

Completely true. The huge problem I see, is that all the classic NLP tagging corpuses are created from the very narrow domain of news articles, and a few good corpuses now appearing for biology texts, and that's about it. Want to do, e.g. NER for product reviews or chat logs? - Incredibly bad results. There's a huge corpus problem in NLP today.

> Why is the competition designed in such a way that it is an advantage

See my comment & links at the bottom of this page. For some reason HN always puts my comments near the bottom, even when they're getting many upvotes. (I'm guessing that HN has a comment-editing penalty from editing my comments too many times, but I'm not sure, or maybe it's from times I must have insulted the mods here.)

Edit:

I constantly get "submitting too fast", so I'll reply here.

Thanks for the info kefka. Yeah, I'll happily agree with that linked comment, that the childish crude moderation techniques used here have been hurting a once great community. It was the exact same scenario for me as for that guy. Seems I got rank-banned immediately after confronting dang about the trends of excessive downvoting and increasing side-project criticizing on HN.

Background:

The technique is actually pretty fascinating. This is something that's been well understood by the cryptography community for decades, but is somehow just recently being fully appreciated by the ML community. See here:

http://blog.mrtz.org/2015/03/09/competition.html

https://www.kaggle.com/c/restaurant-revenue-prediction/forum...

Summary -

Submitting guesses to a system that gives you back scores for your guesses, will quickly leak out enough information that you can reverse engineer a huge number of hidden numbers/labels in surprisingly few iterations, e.g. 700 iterations to covertly extract 10,000+ real numbers with high precision. This surprisingly rapid convergence is a bit reminiscent of the birthday paradox.

Further, this not only lets you win the against the "test" dataset, as apposed to the final "validation" set, but this allows you to significantly increase the data available to you to train your model on, since now you can train your model against both the "test" and "training" datasets.

Layman summary -

ML breaks datasets into 3 partitions "test", "train", and "validation". In cases where they're evenly split, this technique can double the training data you have access to, which is a massive advantage in ML competitions where scores differ by tiny amounts.

Moral judgement -

My opinion, this moral argument is misdirecting the attention from where it needs to be. Yes, it's bad what occurred here. But at this point, in 2015, and with tools readily available to crack this problem effortlessly, it's inexcusable for contests to allow so many scoring reports against their validation sets anymore. It's no longer a question of whether contestants will do it, but how many of them will. We'd might as well just let people self-report their scores on an honor system, if we're going to be this overly trusting.

Try creating a contest system like this in the cryptography field any time in the past 3 decades and you'd be insulted and laughed out. Allowing so many scoring reports against the validation set is fundamentally flawed. The only solution is to globally limit calls to the scoring api.

Another proposed solution -

Allowing everyone to see everyone else's guesses & resulting scores against the "test" set, so that everyone is on equal ground for reverse engineering the "test" set, and then globally limiting the number of scoring attempts so that the test set isn't reverse engineered too significantly.

Overfitting the "validation" set actually is not a problem either way, because none of these contests are dumb enough to let anyone score against the validation set at all until the contest submission deadline is over.

It's always good to have, no matter what, and it would be extremely easy to just say you don't have it if it was a negative or worthless.

The problem is - opportunity cost.

Sounds like some of the worst advice I've ever heard.

Employees being considered property. Disgusting. US universities have become the most prevalent predatory institutions in existence. Earning trillions per year and burdening millions with increasingly crippling debt, while only a minuscule fraction of that money actually goes to the professors and researchers.

Say what you want about Uber, but universities are clearly 1000x morally worse.

Places with the simplest contract law tend to be the worst = whatever is on the contract is what happens. I'm not quite sure what you're trying to say.

Nearly all countries in EU will allow companies to aggressively pursue violators of no-hire and no-solicit clauses. Do you know of anywhere in the EU where the courts automatically reject those types of contract clauses?

I almost get a laugh whenever EU region people say that Americans are slaves to their jobs. Somehow they actually believe the BS that the EU labor laws are vastly superior for employees in every way.

What you're saying is true in many places in the EU. It's very sad.

It's as if the CA anti-solicitation scandal never happened. Let's increase the fines for companies who are trying to own their employees by 100x.

Curious to point out that Tinder & clones found a solution to the dating churn problem:

- Don't target people for whom success in the app will make them want to churn = people looking for a public committed relationship.

- Target people for whom success will make them want to use the app more = people looking for secret private hookups!

Parallel construction.

Historically, doesn't seem to be true at all. Local law enforcement likely cares far more about prosecuting you individually and has far more means to do so, than some foreign national agency. Imagine if Twitter was based within one of those countries that had a major uprising. Would the locals still be as safe using it, as if it were based far beyond the control of their country?

Not really anything unknown about this situation. There are many obvious historical incidents.

Yes, your local government controlling your communication channels will result in far better privacy than some remote government who doesn't care about you controlling it. /s

And the best way to do that is to eliminate the crutch.

Many of the more data-centric fields did away with p-values long ago in favor of far more robust approaches to establishing the value of a paper. Look at papers in the data science field. Using p-values in a paper there would get you laughed out of the industry fast.

Only problem that p-values ever solved was allowing journal reviewers to be lazier, helping scientists to be lazier, and making writing more sensational press releases easier.