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

Faster neural networks straight from JPEG (2018)

uber.com

Post reply on HN
Print view View original

81–90 of 108 posts

Re: Faster neural networks straight from JPEG (2018)

#81

Post by ummonk »

Llamamoe wrote: post #63
whimsicalism wrote: post #61

Earlier quoted context omitted.

In deep ML, people are pretty familiar with the bitter lesson and don't want to waste time on this.

The bitter lesson is about not trying to encode impossible-to-formalize conceptual knowledge, not avoiding data efficiency and the need to scale the model up to ever higher parameter counts. If we followed this logic, we'd be training LLMs on character-level UTF-32 and just letting it figure everything out by itself, while needing two orders of magnitude bigger contexts and parameter counts.

Heck why even go that far? Given how much texts we have in scanned books, just feed it scans of the books and let it dedicate a bunch of layers to learning OCR.

Re: Faster neural networks straight from JPEG (2018)

#83

Post by philipphutterer »

ummonk wrote: post #81
Llamamoe wrote: post #63

Earlier quoted context omitted.

The bitter lesson is about not trying to encode impossible-to-formalize conceptual knowledge, not avoiding data efficiency and the need to scale the model up to ever higher parameter counts. If we followed this logic, we'd be training LLMs on character-level UTF-32 and just letting it figure everything out by itself, while needing two orders of magnitude bigger contexts and parameter counts.

Heck why even go that far? Given how much texts we have in scanned books, just feed it scans of the books and let it dedicate a bunch of layers to learning OCR.

Or given the number of unscanned books, even just give it the controls for a book scanner, the books and probably some robot arms. Then let it figure out the scanning first in some layers. Shouldn't be that hard.

Re: Faster neural networks straight from JPEG (2018)

#84

Post by typon »

ianlevesque wrote: post #11
capableweb wrote: post #9

Earlier quoted context omitted.

Ah, just like I "make hamburgers" when I go through the McDonals drive-through, although it's via a business transaction.

God this website is cynical. Licensing technology from another company to commercialize it is completely legitimate.

They had the Uber ATG group - they laid everyone off and gave up on their AV research. How is that cynical? That's a fact.

Re: Faster neural networks straight from JPEG (2018)

#85

Post by sdenton4 »

xenadu02 wrote: post #70
sdenton4 wrote: post #31

Earlier quoted context omitted.

There are real differences in audio perception, though. Frequency range and sensitivity to different frequencies is a big difference in other animals; I would expect cats (who chase rodents, which often have very high pitched or even ultrasonic vocalizations) to be more sensitive to high frequencies than humans, and thus low passed / low sample rate audio could sounds 'bad.' Another aspect is time resolution. Song bi…

True but hearing is logarithmic in both volume and frequency domains. Double the power does not equate to anything near double the loudness. Similarly each doubling of frequency is only one octave higher. Hearing up to 80khz doesn't mean hearing 4x more than humans... 10 octaves for humans, 12 octaves for cats. In a musical sense it probably isn't noticeable.

The extreme upper limit of human hearing is around 20khz, so cats really are hearing things that we don't, and for good reasons.

Sensitivity to different frequency ranges is more or less independent of anything else. Birds have heightened frequency response in the range they vocalize in, which helps them hear others if their species. Same for us; we vocalize at relatively low frequencies, so most of our hearing ability is focused on that range. There is also a range below which we don't hear: infrasound, which is utilized by elephants.

Logarithmic perception is certainly real, but the tuning of which frequency ranges an animal is more or less sensitive to is certainly species dependent.

Re: Faster neural networks straight from JPEG (2018)

#86

Post by ulrikrasmussen »

The quantization used for JPEG is optimized to throw away information in the frequency space that doesn't matter much to human perception, but I wonder if that is also optimal for training neutral networks?

Also, as far as I know, the human eye doesn't process images in blocks. I also wonder how blockless encoders such as JPEG 2000 would fare in this approach.

Re: Faster neural networks straight from JPEG (2018)

#88

Post by jacobgorm »

I’ve tried this idea in the past and found it not very useful in practice. It breaks when you want to add image augmentation during training, and JPEG is anyway a pretty lousy format for storing training samples if you care about saving space or retaining image quality.

Re: Faster neural networks straight from JPEG (2018)

#89

Post by asah »

fxtentacle wrote: post #56
corysama wrote: post #34

Earlier quoted context omitted.

As an AI armchair quarterback, I've always held the opinion that the image ML space has a counter-productive bias towards not pre-processing images stemming from a mixture of test purity and academic hubris. "Look what this method can learn from raw data completely independently!" makes for a nice paper. So, they stick with sRGB inputs rather than doing basic classic transforms like converting to YUV420. Everyone lea…

In my experience, staying close to the storage format is very useful because it allows the neural network to correctly deal with clipped/saturated values. If your file is saved in sRGB and you train in sRGB, then when something turns to 0 or 255, the AI can handle it as a special case because most likely it was too bright or too dark for your sensor to capture accurately. If you first transform to a different color s…

Would this apply to music formats as well ?

Re: Faster neural networks straight from JPEG (2018)

#90

Post by nyanpasu64 »

xenadu02 wrote: post #70
sdenton4 wrote: post #31

Earlier quoted context omitted.

There are real differences in audio perception, though. Frequency range and sensitivity to different frequencies is a big difference in other animals; I would expect cats (who chase rodents, which often have very high pitched or even ultrasonic vocalizations) to be more sensitive to high frequencies than humans, and thus low passed / low sample rate audio could sounds 'bad.' Another aspect is time resolution. Song bi…

True but hearing is logarithmic in both volume and frequency domains. Double the power does not equate to anything near double the loudness. Similarly each doubling of frequency is only one octave higher. Hearing up to 80khz doesn't mean hearing 4x more than humans... 10 octaves for humans, 12 octaves for cats. In a musical sense it probably isn't noticeable.

As a comparison with removing the top two octaves from a cat's hearing, try removing the top two octaves from an audio file compared to your hearing range (lowpass at 5 kHz or less if you have hearing range loss, and/or resample to 10 kHz/ksps or less) and see if the results are musically noticeable. (At least for humans, the result is intelligible but heavily muffled, I can't speak for my pet cats though.)
Post reply on HN