4 ms·
Your comment seems to be a complete non-sequitur to this article. Buzzword1+Buzzword2 = interesting. Why? What part? What application?
by rubidium 10y ago
Your comment seems to be a complete non-sequitur to this article.
Buzzword1+Buzzword2 = interesting.
Why? What part? What application?
- baq 10y agowith a proper algorithm which may well not exist it's in theory possible to e.g. perform deep learning on all training examples in a single pass or something like that.
- Panoramix 10y agoBut those hopes are not very much based on reality so why entertain them in the first place? It's a bit like saying that QC will be awesome because thanks to it people could in theory find the cure for cancer, in some way, somehow.
- monocasa 10y agoTo be fair, even the non general QCs like D-Wave are supposedly great at protein folding, which is one of the barriers to various cancer cures.
- neom 10y agoI didn't realize imagination was banned from HN now, my bad.
- Dylan16807 10y agoImagination that's not based on anything is generally considered to be off-topic, sure.
- zitterbewegung 10y agoCan you tell me the quantum algorithm that will do this ?
- baq 10y agoi cannot because i'm pretty sure it hasn't been discovered yet and cannot be certain it's even possible to figure one out. IANACSPhD.
- neom 10y agohttps://news.ycombinator.com/item?id=12727836 https://news.ycombinator.com/item?id=12727836
- johncolanduoni 10y agoWhile the OP is wrong about being able to "do everything in parallel" via quantum superposition (we have plenty of no-go theorems when it comes to quantum computers being faster simply due to parallelization) some interesting algorithms that are relevant to data science in general have been developed. For example, this[1] one provides an exponential speedup (over known classical algorithms) for calculating persistent homology, a useful method for identifying scale-independent patterns. [1]: https://arxiv.org/abs/1408.3106 https://arxiv.org/abs/1408.3106