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

There's a lot of research into this. If you're curious you can Google around for transfer learning and one shot learning.


Thanks for the terms to search.



The vast majority of "action" available online is "6-max" (6 player) or "full-ring" (9 player). On a PokerStars, WSOP.com, or Party Poker, you're going to find that there are maybe 1/10th or 1/20th the number of headsup tables as higher capacity tables.

The development of "GTO" (game theory optimal) play in Texas Hold 'Em is certainly a first step in the direction of computers playing poker. However, there's still quite a long way to go.

Poker Snowie, one of the cutting edge "GTO" programs, is based off of NNs (https://www.pokersnowie.com/about/technology-training.html). At the same time, there are some glaring weaknesses in the software, namely that it can only offer suggestions at specific pot size bets (0.25, 0.5, 1, 2). The authors themselves concede some other weaknesses (https://www.pokersnowie.com/about/weaknesses.html).

Worth mentioning that Amaya, the owner of PokerStars, has posted job openings for "AI researchers" (http://www.starsgroup.com/careers/job/Poker-AI-Research-Engi... & https://www.pokernews.com/news/2017/10/pokerstars-to-hire-ar...). Some people think that the position may be to help PokerStars detect/combat bot use, but others think that there may be a (arguably bleak) future where players have the option to compete against Amaya-created bots online.


Here's the paper about Libratus (mentioned in the article).

In January 2017 Libratus beat a team of four top-10 heads- up no-limit specialist professionals in a 120,000-hand Brains vs. AI challenge match over 20 days.

(PDF) https://www.cs.cmu.edu/~noamb/papers/17-IJCAI-Libratus.pdf


They have armed guards. They did fix it. :)


>They have armed guards and nobody else is armed.

FTFY


Great point. Just like Venezuela!



You mean gentrification?


How "modern man" forgets that the comforts of a "middle class level" are unobtainable for all but the Uber wealthy in the world.

Your problems are problems. But they're not real problems. Real problems like starving to death. And not "if I lose my job and x and y and z then I'll starve", but "I'm dying" kind of starve.


Your logical fallacy is: Not as bad as

https://rationalwiki.org/wiki/Not_as_bad_as


Wrong.


Wrong


I think it's funny that people still don't know anything about the case that codified that shouting fire in a crowded theater should be illegal.

Really not the strongest argument when you just give a blase retort with no actual substance.

The corporate media seems to get away with fraud, libel, and slander daily. So probably?


Until you realize over 50% of the school is the progeny of the 1%. Then you realize what's actually going on.


Count dankula has a communist flag tattoo. He marched and demonstrated with socialists.

The fud is weird.


With all do respect, that's not the best deterrent. The best deterrent is the door.

Anyone who knows anything about the Germanwings crash (https://en.wikipedia.org/wiki/Germanwings_Flight_9525) knows this. The pilot used an axe on the door for minutes while knowing he was going to die. He didn't get in. Everyone died.

The risks of hijacking are slim to none, unless you're going to hack the plane. If you're going to hack the plane you just need a laptop.

The TSA is a federal jobs program. A nice joke I heard is to disband the TSA and have them be the workers building the wall. Sure it would be useless, but at least it wouldn't inconvenience law abiding citizens just trying to go about their day, treating them like criminals.


We could in theory draft them into the Army Corps of Engineers and make them push wheelbarrows full of sand and gravel around in the Sonoran Desert.

I like this idea.


While it might just make me feel better about all the times they made me remove my laptop from my bag, or take off my shoes, I would almost rather see them given real skills so they can be a useful part of humanity.


If you’re talking hijacking sure, i still think someone claiming “i have a bomb” will get commands through the door without it ever opening if they’re smart enough about it. The door is a good defence from immediate physical harm for sure, and probably the best change the industry has made in a long time, but i still think the security through stated obscurity element of the TSA could be improved to socially influence those considering any act against an air crew.


So the choice is:

1. Open the door and the person uses the plane as a big bomb.

2. Don't open the door and the person uses a smaller bomb.

You die either way. What's the choice here?


That’s kind of my point? The door while having an impact doesn’t really change much.


> The door while having an impact doesn’t really change much.

What do you mean? Without access to the controls of the plane, they hijackers cannot accomplish their goal of using the plane as a big bomb. They are deterred because it costs a large amount of operational resources to only fail the mission. If they wanted to simply kill people, they have simpler means (much cheaper means) to hit soft targets.

Not only that, the complex, involved missions involve more people which raises the risk of failure, of being caught ahead of time, and of unwinding the clues up the stack. If they are going to take those risks, they want a certain chance of success * size of outcome.


I said this below, but just to throw it here anyway: while that's been true up until now, i'm expecting that to change. Using the plane as a bomb has historically required cockpit access, i expect in the future we'll see a similar attack without cockpit access...just because people will get creative and find a way, if they really want it.


I'm not really at all sure how to even begin to respond to that.

I mean surely you think that flights AA 11, ua 175, and AA 77 were materially different than UA 93. So I really don't get your point.


You mean that one made it to it’s target and the other didn’t? Maybe I’m misunderstanding your point because it doest really feel like a discussion, rather an interview?

Like, sure, if that’s success then ok...again I’m just saying i think there are alternative ways to improve security, and while a locking door is one of them, the TSA is probably another great place to look for improvements. It feels like the US airport security situation is stuck in the world of “look busy and sound strict, people will behave and respect the security of planes”, but really it’s just a giant pain in the ass for most people and meanwhile mostly ineffective at stopping actual risks from passing through to the aircraft.

Real question: do you work for the TSA or airport security in some way?


I mean I feel like it's obviously true that a successful hijacking means using the plane as a bomb.

If you want to just kill a lot of people there are far cheaper and more effective ways. Check out the concerts being attacked.

So the door is all you need.


Yeah, it's certainly a big success to use the plane as a bomb. I'm really just trying to say, and obviously doing a bad job at it, that i don't think access to the cockpit is required to achieve this. So far that hasn't been demonstrated, but the fact that the cockpits were so accessible in the past obviously meant that people didn't need to worry about it. I predict in the future we'll see similar attacks carried out, despite not getting access to the cockpit. It'll require creativity, but it just seems like a matter of time to me.


Consider applying for YC's Summer 2026 batch! Applications are open till May 4

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

Search: