backpropaganda

9 ms·

61.
by backpropaganda 8y ago
Great test. The bots won't be able to pass. Both fast reaction times and lack of cohesive strategy will give them away in the first 5 minutes.
62.
by backpropaganda 8y ago
This is the kind of misinformation OpenAI is so great at. Otherwise smart people like tlb are led to believe that this was the full game even though it's not even close to the full game. Thanks for helping me make my point.
63.
by backpropaganda 8y ago
> The bots were routinely pulling off coordinated team behavior that players couldn't figure out, but that worked. This has to qualify as complicated. This is just not true. Do you know the game? Are you speaking as a player? Or are
64.
by backpropaganda 8y ago
> I think it is amazing that one relatively simple mathematical construct can produce such a complicated behavior. No complicated behavior was produced. > Or, I don’t know, maybe it says something about the complexity of Dota 2 game?
65.
by backpropaganda 8y ago
I think parent worded it in a way that's causing confusion. The point is that in India, tech is one of the only few ways to earn a decent salary. The average American has many more ways to earn a decent salary and thus the choice to go
66.
by backpropaganda 8y ago
They can memorize any supervised learning task, but so far, we haven't been able to see any deep generative model successfully memorize something more complex than MNIST.
67.
by backpropaganda 8y ago
I doubt this AI gets perfect performance even in the training set. Deep generative models are known to underfit more rather than overfit, i.e. they can't even do a good job of the full training set let alone the test set. The cherry-pi
68.
by backpropaganda 8y ago
The team spirit hyperparameter is a crutch they've introduced themselves. Ideally it should be one. In Dota there is only one objective for the entire team and that's to win the game. The fact that they shape rewards is an impleme
69.
by backpropaganda 8y ago
That's not the point. The point is that the humans wouldn't know that regen-heavy deathball strat is optimum unless they've played that game a few times. Since multi-invincible-courier game was not available to the casters, t
70.
by backpropaganda 8y ago
Axe's cast time (also called cast point) is 400ms. Source: https://liquipedia.net/dota2/Axe This means that Dota 2 was balanced around the expectation that human's best reaction would be at least 400ms, becau
71.
by backpropaganda 8y ago
There are many things you could do to make the bots not fight for courier. Just implement a queuing system with a priority for cores. Or better yet, make the couriers not usable if another hero is using it. The way I would guess this played
72.
by backpropaganda 8y ago
It's not independent agents. The neural networks have the same input, share weights, and also share some activations. With that much sharing, it's better to think of it as one neural network which has output heads for all the 5 pl
73.
by backpropaganda 8y ago
> And we still don't have language models that make any sense at all. My understanding is that current language models are very good at capturing the syntactic structure of the sentences, but fail when modeling would require common
74.
by backpropaganda 8y ago
Not sure if you're trolling but a tool which flips between two popular genders doesn't deny existence of other genders. Not everything related to gender has to be about the gender spectrum.
75.
by backpropaganda 8y ago
Hence the question: Why is it considered good UX to make your readers switch context to do a google search and then return to reading the article?
76.
by backpropaganda 8y ago
Is the average reader of NYTimes really so well-versed with Latin that they'd know phrases such as "sine qua non"? I've typically seen such phrases used when the context makes the meaning clear, but this was a real stump
77.
by backpropaganda 8y ago
You're missing the part where this could happen to anyone including you on any regular day. The way we've set our systems and laws of probability up, you might see yourself handcuffed a random day of your life and go through a sig
78.
by backpropaganda 8y ago
> an airport that is specifically designed to make everything as strait forward as possible You'll find that you're in a very very small minority of people who think airports are designed to make everything as straightforward a
79.
by backpropaganda 8y ago
I'm curious to know if the AI can win with a 50-50 draft. It can be argued that the main reason the humans lost was because they didn't have a feel for the meta of this 18-hero version of Dota, while the bots did. What if you give
80.
by backpropaganda 8y ago
From [1], the teams alternatively pick a hero from the pool. In this case the pool is not random, but fixed to the 18 heroes. Calling this Random Draft is confusing. Would be nice if they can confirm that the humans would have a choice in t
81.
by backpropaganda 8y ago
Yes, the reward for a team includes a term which is negative of the enemies rewards. This was mentioned in the previous blog post. Here it is: https://gist.github.com/dfarhi/66ec9d760ae0c49a5c492c9fae939...
82.
by backpropaganda 8y ago
Using certain names and terms on Reddit and Twitter (not sure about HN) does bring bots in who then try to influence the conversation with votes and propaganda comments. So J. K. Rowling made quite an accurate prediction about how society m
83.
by backpropaganda 8y ago
Dividing fractions is also easy. Did you mean summing fractions? Because, yeah, summing fractions is significantly harder than multiplying fractions, simply due to pure misfortune.
84.
by backpropaganda 8y ago
Funny how when I replaced the big hyphenated word with the 4-letter name, it actually triggered an emotional response in me which made me dismiss the rest of your comment, even though I was fine with the comment without the substitution. We
85.
by backpropaganda 8y ago
You need to pick up linear algebra and basic probability. CS229's linear algebra review is pretty good: http://cs229.stanford.edu/section/cs229-linalg.pdf For more depth, Trefthen and Bau's Numerical Linear A
86.
by backpropaganda 8y ago
What is your math experience/background? In any case, for the average case, this course is probably the best for those who want a slightly deeper understanding of machine learning than offered by Andrew Ng's courses: https:/
87.
by backpropaganda 8y ago
I'm not saying I approve of this. Just stating how things are. I'm not an interviewer and don't represent Google, but I doubt other big companies have any better process.
88.
by backpropaganda 8y ago
> I barely take this into account. > I file feedback exclusively on technical merit and communication. I appreciate your comment, but my point is that there's no way to actually do this, even if your goal is to. It may seem that
89.
by backpropaganda 8y ago
The strongest factor in my opinion (after pure luck) is the rapport you're able to build with the interviewer. Doing that of course requires doing well in the technical questions, but the end-goal shouldn't be seen as doing well i
90.
by backpropaganda 8y ago
I wasn't as convinced by the mathiness and speculation/explanation arguments as the others. Sure there may be a few examples of obfuscation via math, but most papers, in my view, don't do that. Adding the couple of math equat

More ›