The Neuralink Compression Challenge wants your help
Elon Musk’s Neuralink recently succeeded in implanting its first brain chip in a real-life patient. However, the company has a compression problem, and it invites the public to help.
The brain implant is a technological marvel that turns one’s thoughts into computer inputs. As a result, Neuralink intended it for various physical disabilities.
READ: Neuralink brain implant encounters problem
Article continues after this advertisementThe current version of the neural implant takes too long to transmit data. That is why the Neuralink Compression Challenge asks the public for a new algorithm to solve this problem.
What is the Neuralink Compression Challenge?
I took a look at the Neuralink compression challenge.
— Casey Handmer, PhD (@CJHandmer) May 28, 2024
The data has a bunch of noise, so 200x compression isn't possible, at least according to any method I'm aware of.
But if you truncate to 4 bits you cut off the noise and if you subsample from 20 kHz to about 400 Hz (I used… pic.twitter.com/lLrCk6PzSb
The first thing you’ll notice about the Neuralink Compression Challenge website is it is highly basic. It only uses HTML, meaning it only has text and no graphics.
Nevertheless, the page explains the N1 brain chip implant generates approximately 200Mbps of electrode data and can transmit around 1Mbps wirelessly.
Article continues after this advertisementThat is why it says it needs over 200x compression. Moreover, it must run in real-time in less than a millisecond and less than 10mW.
Engineers and other experts may send their algorithms that can solve the problem. The Neuralink Compression Challenge will score submissions on the compression ratios they achieve on different sets of electrode recordings.
Canada-based news firm CBC reported the social media response to the challenge. Some experts said it was “impossible” because Neuralink staff wanted to convince Elon Musk that it couldn’t be done.
Karl Martin, chief technology officer of the data science company Integrate.ai, is one of these skeptics. He explained to CBC in an email that Neuralink brainwave signals are compressible around 2 to 1 and up to 7 to 1.
However, 200 to 1 “is far beyond what we expect to be the fundamental limit of possibility.” Also, earlier versions require bulky wires to communicate with an external device.
That is impractical in the real world, so companies are developing wireless brain implants. Neuralink believes compression is the key to this development.
Martin said that solving the Neuralink Compression Challenge requires going beyond the laws of physics. In other words, he says “It’s essentially seeking a miracle.”
