Hacker News Reader

Comment by emilwallner | Hacker News Reader

While the results are really cool, they're not perfect, and seemingly even the best companies working with deep learning are still not getting it right every time. It makes me wonder if this technology will ever be useful for these tasks... I mean would you be willing to use BCIs even if they made small errors from time to time?

View on HN

There is a lot of variance in error rate in biological human nerves as well. Certainly a lot of people on the low-end of the bell curve could benefit.

Interesting to consider whether a 'filter' could be made for people with parkinson's that would drop tremor noise while preserving intentional actions. Perhaps with robotic exosleeves tremor could be offset/counteracted - or eventually we could input the filtered signal back into the nerve.

That would be a fantastic application, however it would differ slightly in that it would not be a BCI (as in a traditional brain computer interface), but a device within the brain that impeded certain signals being sent to motor neurons that it detected were Parkinsonian. This would involve directly manipulating the output of the brain to the body itself, and require some serious bio-mechanical hardware!

This hardware already exists as an implantable electrode and is used in really hard cases.

It is a competitive inhibitor and additionally stimulates dopamine release.

As for putting something like that in spinal cord or motor centers, very risky surgery. Much more than cranial stimulator one. If this device fails in such a simple way as being jostled, you die.

I think being able to get 90%+ accuracy is pretty bad ass. I think with more data over time, there can hopefully be a good amount of progress that can be made.

I don't even know what would make sense to improve the model (such as lowering false positive or false negative rates), but this research sounds pretty exciting!

See the progression of voice recognition. The answer is, yes, a BCI is still useful and will still be used, even if it makes errors. It will of course grow more useful and become more embedded into our lives, as it gets more accurate.

Yes that's the big question, and I mention that in the article. Let's say for a moment that we can't achieve 100% accuracy, well one parallel to this situation would be autocorrect on our phones; it often corrects you when you didn't want it to or misinterprets your meaning. Yet we still use it as there is net gain in terms of convenience. If a BCI could instantly put your thought on to paper, and left you solely with a few typos to mop up, I think we'd all be willing to compromise.

Reply on news.ycombinator.com