devilmoon

It's interesting to me that most comments are confusing Deep Learning with the whole field of AI in general, when it is actually a subset of Machine Learning. Well deserved anyhow!

4% for each European user, theoretically. Realistically, nothing is gonna happen and they will keep happily earning a fuckton of money from unaware users.

I can actually log into Facebook with a password I used on there roughly two years ago, which to me seems like a VERY bad security issue. I'm hoping they are at least fingerprinting my devices so that if someone else tries to access my account with an old password it won't let them, but still...

Even if not pre-rendered this doesn't seem _that_ good to me to be honest. Some of the things I've seen in BF5 with Ray Tracing were more impressive to me than this whole demo. i.e. when the fairies fly away the lightning on the water seems really off to me, even when they fly around the crown/princess it looks more weird than stunning.

I don't know, there doesn't seem to be any mention either way, if it's not pre-rendered it is slightly better but still, I don't understand why a packed room at GDC of all places would be wowed by this.

Am I missing something? This doesn't really seem all that great compared to what we have been shown with Ray Tracing. The fact that it's not even real time but pre-rendered makes it even less wow-ing for me

Something I have noticed throughout my studies is that some countries seem better equipped to let university students shine and not suffer from high stress, or at least help them decompress.

This might be anecdotal experience, and I am not currently in a PhD programme so I am not sure it would aplly there as well, however this is what I have noticed:

In my country universities are usually an environment quite similar to High School wrt how students usually fit into the greater social circle of the institution, meaning that you usually go in for your lectures and leave as soon as you're done, have only a few ties with other students and there are almost no extra curricular activites to partake in through which you could bond with other people. During my undergraduate studies I spent a year in a university in the UK and this experience opened my eyes to how detrimental this environment is to students' performance; The university I attended explicitely encouraged students to set up "Student Associations" centered around whatever topic the person setting up particularly cared about, and in my experience I attendend events from many different ones, ranging from my department's one to anime/manga, to gaming, to my native country's culture. This has helped me so much in connecting with other people I would've otherwise never met who shared some of my interests, and in turn I was always able to have someone to speak to about common problems that students face, or to attend social events in order to decompress from a particularly hard day; at the end of the year I gave some of the best exams of my career, I made friends I am still in contact with years after, I was able to land an internship through one of these societies' events, and generally enjoyed life and university MUCH more than I ever did in my country. After coming back home and moving on with my life I am currently pursuing a Masters in my own country, and after a bit more than a semester I already feel the same feeling of dread I felt during my undergrad experience kick in since there are no avenues or ways to actually enjoy life as a university student.

I firmly believe that if more countries / universities adopted the same stance I experience in that UK university students would generally be way better off, both with respect to their mental health and their track record.

Unfortunately I have never met someone who has shared the same experience as me and hence has the same outlook on the problem, and even when speaking up about these issues with my department or colleagues many seem to not particularly care or understand what that particular system would entail for everyone involved. It is really a shame.

I am wondering if there are any scientific studies that have been done on this particular issue that I could bring up to shift perception around this?

Anyway, just my two cents.

tl;dr: make students more involved in their university's community and provide the means to self-organise around common interests so as to create comraderie and avenues to decompress, everyone will benefit from it.

Just skimmed through it but the main thesis seems blown out of proportion IMHO; Google still works very well (read: better than all its competitors), what bugs me a bit is that it tailors result to my previous history hence if I look for something that is different from what Google is expecting from me results won't be _that_ good. There's also to consider the fact that people have learned how to game the algorithm through SEO techniques and that's not a problem with Google itself but it something that could be applied to every search engine (at least with the underlying Information Retrieval theory that we have now).

Hey dude, Italian here as well currently enrolled in a Masters. I am thinking over the possibility of going forward with a PhD rather than looking for a job, could I ask you some questions? It is quite hard for me to find actual information on PhD life here in IT and unfortunately I don't know anyone who has done one.

Meaning that the final models evolved by playing against each other and they all started by using pro strategies, so, again, to me it seems kind of obvious that they would end up using pro strategies and in the best case just try to make them better

Well, why wasn't this implemented then? For an hyped and live streamed event I'm expecting to see something that blows me away, not hand picked agents still in embryonic stage

I do believe the live match was with a newer version of the agent that hadn't been tested against human yet

>I think the coolest thing is how the play of AI mirrors a similar style to how pros have developed

I mean, the agents have learned the game from pro replays so to me it seems obvious it would evolve to play the same strats human pros use.

I'm not proficient in Starcraft, but if they have to make Pros play against 5 different agents for a Bo5 is it because the same agent would merely repeat the same game overall and the human player would be able to see through its strategy?

AFAIK at least for Go/Chess DeepMind wasn't handpicking agents to send against human opponents, but it was simply a trained agent who would try its own strategy and respond to the opponent, in this case isn't it like the single agent only ever plays a single type of meta strat? If so, I think this is a bit less impressive than what I had predicted.

Holy shit, I remember playing something very similar to this for hours on end in the early 90s as a kid; it might have even been an early version of the powder toy (another user mentioned it), not entirely sure, but regardless this thread is making me go back down memory lane and now I'd love to get into the insides of a game like this one and understand how it works!

Thanks for your input, I definitely want to set it up asap so that I can start blogging about stuff I'm interested in or ramblings I want to go on about after tackling some of the bigger assignments.

I feel like the problem lies in getting your feet through the front door, after that how you got your education really doesn't matter anymore. You were lucky enough to get an interview with Microsoft thanks to your past work experience, but other people (especially the ones who try academia before going into the private sector) don't have that chance and most of the time won't even be invited to an interview, that's the problem.

I am confident that if big companies actually interviewed people coming from different universities they would find that most top-uni grads are worse that people coming from different institutions who always feel like they have to perform 110% just to level the playing field.

I did mean FAANG companies, yeah. At the same time I feel what s/he's saying is also right; in fact I am slowly building up repos on github while I work on my masters and hopefully I'll start contributing to librabries I use so that I can show all of that off to potential employers. I also feel like having a website in your name works wonders, but I am not too sure companies are really interested in that.

It matters especially if you're not from the US. You could've gone to the best university in your country, but if it's not a top institution in the US or a handful of others (ETH and a few more in top positions according to Qs Rankings) I feel like most big names in tech won't even look at your resume.

Anedoctal example, but I graduated with a decent mark from the top CS department in my country (106/110) as an UG, I couldn't even apply for a masters in top universities in the UK (let alone US) since their requirements are skewed towards not accepting people from my country (if you came from the UK you needed a 2:1, which if I remember correctly is around 70%, if you came from my country you needed 109/110, which is over 99%. Scores for other EU countries like Spain and France were converted to their 70%). This led me to going into a masters in my country, which will most likely hinder me from getting a PhD in any top institution in the EU just because of the country I was born in, even though some of the professors in my current masters degree course are top researchers in their fields. So yes, I feel like where you go to school earlier in life greatly influences where you will end up in the future; and even more so, where you were born influences your chances of making it anywhere in life. This is not to say that I am some unrecognised genius, I feel like I am a "median" CS student with a poor math background to booth; but at the same time I absolutely feel like a lot of talent will go unrecognised and won't have the same chances in life others have just because they happened to be born in a country rather than another (even more so after having studied in the UK for a year and having met people who went on to work for top companies when they clearly were subpar compared to people studying in my own country)

Maybe there is a shortage because wages are low thus people are not programmatically attracted to the field? ;)

I believe it would be quite ridiculous as well, any elected employee representative would probably be somehow bought out by the company/board and become useless. That's what Union Reps in my country do at least - they are basically friends with upper management/board and just strike deals that benefit them rather than the workers

Depends on the website usually, one that I managed to "hack" is a newspaper in my country (the biggest one actually) that only hides the article text through a CSS class; You just remove that from the HTML through dev tools and have access to the whole thing. I hope in most cases it is more complicated than that

Not my personal experience, but one of my professors is trying to get his patch approved into Postgres for an upcoming release; since his changes are very deep and touch the kernel (he wants to add support for temporal data) he told me the whole process will take years and the whole community will examine these changes before anything is added.

I think that pretty much explains it: Cutting edge features are proposed way earlier than commercial DBs start thinking about implementing them, and the community hones them over years before they are released to the mainstream public.

Am I missing something or is 500$ an awful lot for an email server? You could probably set up a personal e-mail server with a Raspberry and have more or less the same functionalities.

Yeah I meant that I loved the sentence! I worked in corporate before deciding to go back to Uni and it really resonated with me, even though unfortunately I wasn't able to already buy my own freedom to not be in corporate but need to be supported by family still. Hope I can make it one day :)

>I firmly believe that money is to buy things that matter to you. To me, not working at a corporation matters.

Damn dude that is a very good way of putting it

I'd prefer maintaining a 10k LoC python project rather than a 10k LoC Java project (which would likely be 8k lines of boilerplate anyway)

I'm guessing they either left or are millionaires many times over by now and don't surf HN

Of course mobile requires less power, but they mention Moss among the launch titles which is in no way a mobile experience. I'm thinking they are going to port and downgrade everything then?

Universities in my country only offer what are called "collaboration bursaries", which is 120hrs of unskilled work paid 10€/h. I already plan on applying, but as you can imagine it is nothing more than a stopgap. I've thought about posting on the schools bulletin board that I'm willing to help people with their PCs and whatnot, or maybe even do tutoring in CS, but it seems like a long shot for someone to actually call you.

As for languages I think that's most likely a no-go, my university is trilingual and most student are required extensive knowledge of the languages to graduate (for which the university itself offers many many MANY courses)