1
Not Only Does It Convert Back Losslessly It Also Has The Noise Removed

Not Only Does It Convert Back Losslessly, It Also Has The Noise Removed

Updated Jun 01, 2024 at 10:36AM EDT by Rebecca Rhodes.

Added May 30, 2024 at 07:18AM EDT by sakshi.

PROTIP: Press 'i' to view the image gallery, 'v' to view the video gallery, or 'r' to view a random entry.

This submission is currently being researched & evaluated!

You can help confirm this entry by contributing facts, media, and other evidence of notability and mutation.

About

Not Only Does It Convert Back Losslessly, It Also Has The Noise Removed refers to memes and jokes about X (Twitter) user @lookoutitsbbear a.k.a. moderate rock's attempt at cracking Neuralink's "Compression Challenge," issued on 18th May, 2024. Despite complaints from computer scientists who deemed Neuralink's data to be "too noisy" to compress accurately, moderate rock claimed to have found a way to compress a file in a manner that "not only" converts it "back losslessly," but also "has the noise removed." Since it is difficult to distinguish data from noise in files this large, jokes about @lookoutitsbbear's claim began making the rounds in May 2024.

Origin

On May 18th, 2024, Elon Musk's Neuralink Twitter / X[1] account posted an open entry "Compression Challenge" asking people to try their hand at compressing "one hour of raw electrode recordings from a Neuralink implant."

However, the challenge appeared to be asking people to losslessly compress extremely noisey files to 1/200th their original size (seen below, left).

According to various internet users versed in computer science, such as @felix_red_panda on X[2] this task is nearly impossible (seen below, right).


Neuralink Compression Challenge content.neuralink.com/compression-challenge/data.zip is one hour of raw electrode recordings from a Neuralink implant. This Neuralink is implanted in the motor cortex of a non-human primate, and recordings were made while playing a video game, like this. Compression is essential: N1 implant generates ~200Mbps of eletrode data (1024 electrodes @ 20kHz, 10b resolution) and can transmit ~1Mbps wirelessly. So > 200x compression is needed. Compression must run in real time (< 1ms) at low power (< 10mW, including radio). Neuralink is looking for new approaches to this compression problem, and exceptional engineers to work on it. If you have a solution, email compression@neuralink.com Leaderboard ↓ d Compression Compressed Name encode decode ratio size size size zip P 2. 2.2 63M 231K 480K Felix @felix_red_panda Neuralink started a compression challenge. They're asking for people to find methods to losslessly (!!) compress files to 1/200th (!) the size but the files are extremely noisy so thats certainly not possible (spectrogram of one randomly chosen file for illustration) hallenge > is one hour of raw electrode reco on-human primate, and recordings Mbps of eletrode data (1024 electrod Linear Spectrogram ... ver (<10mW, including radio). ession problem, and exceptional eng om • • 11:28 AM ⚫ May 24, 2024 4.3M Views 200 300 400 Time ዳ

While some internet users interpreted the challenge as Neuralink employees' attempt to convince that compressing such noisey data losslessly is impossible,[3] other users tried to find a theoretical maximum compression size for the data.

On May 27th, 2024, X[4] (formerly Twitter) user @lookoutitsbbear a.k.a. moderate rock replied to a tweet, claiming to have compressed a file down to 47 KB, adding, "not only does it convert back losslessly it also has the noise removed." The user went on to defend his claim in the replies to his post, saying that noise isn't data, and he can distinguish the two from "experience."


moderate rock @lookoutitsbbear May 27 i got 4.1 5 URL:Winamp Comm 2717 i@usrbinishan · May 27 This converts back losslessly? 1 moderate rock 272 @lookoutitsbbear 514 ılı 342K ... 922 ılı 349K ☐ 1 Follow not only does it convert back losslessly it also has the noise removed 1:21 AM May 27, 2024 4.3M Views LIVING WE Leon Wilk @Sharfpang May 27 How certain are you that you distinguish data from noise 100% correctly? 6 173 1.5K moderate rock @lookoutitsbbear ⚫ May 27 experience ❤ 31 Ill 183K 1757 ❤ 1K III 166K ☐ 1 Show replies • YISTECH @theRealYISTECH ⚫ May 27 Dude noise is still data 3 27 1.1K ılı 97K moderate rock @lookoutitsbbear. May 27 no it's noise 6 27 Show replies Joshie so, not lossless. 5 1712 371 ılı 83K @phys_ballsocket. May 27 7.4K Ilil 320K moderate rock @lookoutitsbbear. May 27 the data is lossless the noise is removed 19 1726 795 Il 305K Show replies ...

The user went on to post a thread defending his compression, saying that the noise in the data stems from the way it was recorded and stored by Neuralink hardware, and that the challenge is set up for failure if taken at face value.[5]

Spread

Various other X (formerly Twitter) users began sharing memes and jokes about @lookoutitsbbear a.k.a. moderate rock's claim, making jokes about how "lossless" conversion is possible if one just disregards noise. On May 27th, 2024, X[6][7] users @erntha and @jothwip posted jokes about @lookoutitsbbear a.k.a. moderate rock's post, gathering over 40,000 and 11,000 likes respectively (seen below, right).


Vintage 1994 Dudemaster Can you explain Lupa's biggest hater Today at 05:53 "A German, an American and a Russian enter a competition to build the best compression algorithm. The German, after many days of hard work, presents his algorithm. It's able to compress pure noise down to a 20th of its original size, and it does so losslessly. The solution is a genius mathematical formula that gets named after him. The American's algorithm is just as good. By using a combination of two existing algorithms, it can compress the noise and to a much smaller size. The Russian presents his solution. It shrinks the original file down to a single byte. The scientific community is shocked. Such feats are impossible! They beg him to explain his algorithm. The Russian says: 'No way will I ever give up the secret of my perfect algorithm. Not only does it compress it losslessly, it also removes the noise"" Work Smart, Not Hard where cube?

Also on May 27th, X[8] user @ziegenhainy reposted @lookoutitsbbear alongside an image of a colorful pixelated square, and an all red square, gathering over 14,000 likes in three days (seen below, left).

On May 29th, X[9] user @colin_fraser posted an image of the botched restoration of the painting "Ecce Homo" alongside @lookoutitsbbear's catchphrase, gathering over 30,000 likes in a day (seen below, right).


Ziege @ziegenhainy look at my compression algorithm, the image on the left uses a ton of data but the image on the right only uses 3 bytes! not only is this compression lossless, I removed all the noise as well! moderate rock @lookoutitsbbear ⚫ May 27 Replying to @usrbinishan not only does it convert back losslessly it also has the noise removed 4:26 AM May 27, 2024 1.5M Views Colin Fraser | @colin-fraser.net on bsky @colin_fraser Not only does it convert back losslessly but it also has the noise removed 4:25 AM ⚫ May 29, 2024 691.4K Views

Various Examples


it's noise nooo the eval script uses diff! it has to be lossless!!!! 14% 0.1% 2% 10 score 55 70 imgflip.com it's noise, 34% 34% the eval script is wrong 14% 2% 959 0.1% 55 85 100 115 130 145 YOU'RE LAUGHING. MY LOSSLESS COMPRESSION REMOVES NOISE AND YOU'RE LAUGHING shadow word: bunny @kayaempress >looking for a new compression algorithm >ask the programmer if their algorithm is lossless or lossy >she doesnt understand |>pull out illustrated diagram explaing what is lossless and what is lossy >she laughs and says "it's a good algorithm sir" >invest >its lossy moderate rock @lookoutitsbbear. May 27 Replying to @usrbinishan not only does it convert back losslessly it also has the noise removed • 1:23 PM ⚫ May 27, 2024 199.5K Views > lossless 4.1x compression > looks inside > lossy Èlia Έ @guineu_roja torvalds/linux <>Code Pull requests 320 Actions Projects 0 Security l Insights this code very faster #447 Open alexozer wants to merge 2 commits into torvalds: master from unknown repository Conversation 19 --Commits 2 Checks 0 Files changed co . alexozer commented on Aug 13, 2017 • edited prepend // to every line greatly improving compilation performance 21 12 74 this code very faster 1 ✪ 1 00 2 KernelPRBot commented on Aug 13, 2017 Hi @alexozer! Thanks for your contribution to the Linux kernel! moderate rock @lookoutitsbbear ⚫ May 27 Replying to @usrbinishan not only does it convert back losslessly it also has the noise removed • 3:22 AM May 28, 2024 402.5K Views

Search Interest

Unavailable.

External References

[1] Neuralink Compression Challenge – READ ME

[2] X – felix_red_panda

[3] X – typedfemale

[4]  X – lookoutitsbbear

[5] X – lookoutitsbbear

[6] X – erntha

[7] X – jothwip

[8] X – ziegenhainy

[9] X – colin_fraser

Recent Videos

There are no videos currently available.

Recent Images 15 total


Top Comments

0000000000000000000000000
Delete
Sensitive
0000000000000000000000000

in reply to djdyyz

I think that illustrates exactly why Neuralink is out over their skis.

Does the compression need to be lossless? They don't fucking know. No one fucking knows.

If they're applying the same rationale that the AI companies are using – just throw everything in a black box with "Markov Chains!" written on the side of it and call it job's a goodun while dusting off your hands before rolling around in your endless stream of investor capital – sure, go lossy and unleash some more fresh hell of half conscious intelligences that even more convincingly appear to feel pain and make free will decisions (even though they don't).

But if the end goal is the preservation of human consciousness itself….well yeah I guess you'd want it to be lossless….right? …..RIGHT???? The fact that they don't even know….and can't even begin to find out yet….means they're just chasing their own tails until hardware catches up with their ambitions. And there's no guarantee it ever will – a resolution of a billion bits is still digital, we'll never know what we don't know we're not encoding.

+9
Reply

+ Add a Comment

Comments (11)


Display Comments

Add a Comment


O HAI! You must login or signup first!