Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

What a Deep Neural Network thinks about selfies

karpathy.github.io

Post reply on HN
Print view View original

41–50 of 56 posts

Re: What a Deep Neural Network thinks about selfies

#41

Post by visarga »

the8472 wrote: post #27
nightpool wrote: post #16

>Be female. Women are consistently ranked higher than men. In particular, notice that there is not a single guy in the top 100. This sounds true, but it can't be the real reason—selfies are ranked relative to the other images by the same user . So unless users are taking a lot of #selfies of people of different genders, we can assume the dataset is already controlled for the gender of the person in the image, no? Unl…

Well, one of the other factors is long hair and the tendency to oversaturate the face. Those factors don't seem independent to me, men are less likely to sport long hair and they're also less likely to oversature the face to measure up to some skin perfection standards (think of it as the photographic equivalent of makeup). > but it can't be the real reason Can't? Ontop of the above-listed aspects it is entirely poss…

> focus group testing for computer voices tends to result in female voices being chosen

I personally prefer the Alex voice from Mac OS to female voices. It has nice intonation. If only I could make it correct some of the mistakes it makes, for example not being able to distinguish "read" in past tense from "read" in present tense which makes it sound silly. Another error it makes is confusing "live" as in "live concert" with "live" as in "live in USA" (they are called heteronyms and are a special case in TTS).

Re: What a Deep Neural Network thinks about selfies

#42

Post by visarga »

goodJobWalrus wrote: post #15
falcolas wrote: post #10

Earlier quoted context omitted.

Makes a bit of sense, in combination with the "be female" advice, cutting off the forehead puts the center of the photograph closer to her cleavage, and typically shows off her entire chest.

Cleavage does not feature a lot in the top 100 actually, but I'm half way there, in a sense that I'm a female. I'll definitely try the half-forehead thing next time!

We could try and see if the activation for good selfies comes from the cleavage or the eyes.

Re: What a Deep Neural Network thinks about selfies

#44

Post by yakult »

visarga wrote: post #41
the8472 wrote: post #27

Earlier quoted context omitted.

Well, one of the other factors is long hair and the tendency to oversaturate the face. Those factors don't seem independent to me, men are less likely to sport long hair and they're also less likely to oversature the face to measure up to some skin perfection standards (think of it as the photographic equivalent of makeup). > but it can't be the real reason Can't? Ontop of the above-listed aspects it is entirely poss…

> focus group testing for computer voices tends to result in female voices being chosen I personally prefer the Alex voice from Mac OS to female voices. It has nice intonation. If only I could make it correct some of the mistakes it makes, for example not being able to distinguish "read" in past tense from "read" in present tense which makes it sound silly. Another error it makes is confusing "live" as in "live conce…

You can fix this by misspelling your input text. Use 'red' as the read past tense. Use 'laif' and 'lif' in the latter.

Re: What a Deep Neural Network thinks about selfies

#45

Post by bakhy »

lqdc13 wrote: post #4

A guide on how to take a good selfie that others will like: be female be blonde be attractive Incidentally, Christian Rudder did a really good "study" on the dating site pictures a few years ago: http://blog.okcupid.com/index.php/dont-be-ugly-by-accident/

apparently, she should be white too.

Re: What a Deep Neural Network thinks about selfies

#47

Post by simula67 »

karpathy wrote: post #35
nightpool wrote: post #31

Earlier quoted context omitted.

Neither of those explain why it would rank above the average of other female faces , in general. Discussion about this with the author reveals that I was misinterpreting how they were collecting averages. I was assuming the "like" count was coming from each photo collected, but instead they collected the photos and average likes in individual steps, where the average likes were across recent posts by that user, rathe…

I screwed up on this point by the way - I had done this part of the experiment a few months ago and I incorrectly remembered the details. I went back and looked through the code and adjusted the post with more regarding this important point. In particular: "Now it is time to decide which ones of those selfies are good or bad. Intuitively, we want to calculate a proxy for how many people have seen the selfie, and then…

Still no men in the top 100 ? There must be something deep to learn about the difference in sexes there, I am just not sure what it is.

Re: What a Deep Neural Network thinks about selfies

#48

Post by netheril96 »

nl wrote: post #37
netheril96 wrote: post #14

One caveat with these machine inspired knowledge: they are prone to error, probably more than humans, at least for now. For example, if you train a CNN directly with human faces, its recognition rate comes way below what a human is capable of. Only after you apply tons of handcrafted optimizations, which are mostly black art, will you get close to or surpass a human's capability. Without much domain specific tuning,…

This is more wrong than right. The example is correct, but not for the reasons stated. Humans are very, very good at face recognition . However, CNNs are pretty close to human performance for face detection. Only after you apply tons of handcrafted optimizations, which are mostly black art, will you get close to or surpass a human's capability. Without much domain specific tuning, an AI's insight is far from reliable…

A GoogleNet or VGGNet has tons of parameters. How many convolutional layers are stacked together, the size and stride of each one, where to put the dropout layers, where to put the full connection layers, how they are connected together, global learning rate and momentum and decay, local learning rate and momentum and decay, each of these myriad parameters have an unpredictable effect on the final result. The initialization of the network also has a major bearing on the final outcome. It is almost a chaotic system where nothing small can be safely ignored. One time my result of training a CNN was swung by the `batch_size` parameter and to this day I don't know how.

Those parameters are exactly the type of handcrafted optimizations I am talking about. You cannot just fill in arbitrary numbers and expect the network to fare well. In fact, you cannot even expect it to converge.

You can take those papers and build a world class classifier only because someone else has taken all the time to optimize for the specific case. Once you switch the task, the result will be OK, but nowhere close to what a human or a true AI would give you. Not until you take the time to optimize the parameters.

Re: What a Deep Neural Network thinks about selfies

#49

Post by gus_massa »

lqdc13 wrote: post #4

A guide on how to take a good selfie that others will like: be female be blonde be attractive Incidentally, Christian Rudder did a really good "study" on the dating site pictures a few years ago: http://blog.okcupid.com/index.php/dont-be-ugly-by-accident/

Also, long hair in front of your shoulders (no ponytail).

Re: What a Deep Neural Network thinks about selfies

#50

Post by nl »

netheril96 wrote: post #48
nl wrote: post #37

Earlier quoted context omitted.

This is more wrong than right. The example is correct, but not for the reasons stated. Humans are very, very good at face recognition . However, CNNs are pretty close to human performance for face detection. Only after you apply tons of handcrafted optimizations, which are mostly black art, will you get close to or surpass a human's capability. Without much domain specific tuning, an AI's insight is far from reliable…

A GoogleNet or VGGNet has tons of parameters. How many convolutional layers are stacked together, the size and stride of each one, where to put the dropout layers, where to put the full connection layers, how they are connected together, global learning rate and momentum and decay, local learning rate and momentum and decay, each of these myriad parameters have an unpredictable effect on the final result. The initial…

A GoogleNet or VGGNet has tons of parameters.

Kinda, but they are defined for you. For example the GoogLeNet design is described in[1]. Page 5 lists the parameters, the diagram on page 6 shows how the layers are linked.

Yes, I agree that the design of a new neural network architecture is a skilled process, and there is a lot of hard work there. I couldn't agree with that more, but that isn't what we are talking about here.

It is quite possible to take a CNN like GoogLeNet designed for a specific purpose and reuse it in similar situations. GoogLeNet will always do pretty well for image classification.

I think of it as analogous to a piece of software like a database. Designing a new database system is hard, but taking something like SQLite and using it is easy. Yes, you can tune it and get better performance out of it, and yes, it will break if you use it in the wrong circumstances, but it is generally pretty reliable if used as designed.

Now this analogy breaks down because industrial use of CNNs is pretty new compared to Database systems. It's more like trying to get msql running on your Slackware 0.9 system in 1993 it is getting Postgres on Ubuntu 15.10.

Nevertheless, there isn't really a black art to using an existing CNN. Lots of schlepping to get CUDA running on your machine, though.

[1] http://www.cv-foundation.org/openaccess/content_cvpr_2015/pa...

[2] Not MySQL, msql: https://en.wikipedia.org/wiki/MSQL

Post reply on HN