3 ms·
Great write up on the state of the field, but when I checked last, the current problem is performance. I didn't see much on that in the article. A few years ag
by awesomeMilou 4y ago
Great write up on the state of the field, but when I checked last, the current problem is performance. I didn't see much on that in the article.
A few years ago there were papers on evaluating simple logic circuits in an FHE context and it took 2h hours for what was basicially 5-6 NOR gates.
- johnbender 4y agoFwiw we have at least some reason to hope in this general context that between clever systems work and tightening theoretical bounds via additional assumptions and clever reasoning we might get to practical implementations for some applications. As (maybe weak) evidence the progress on practical implementations of PCPs/SNARGs https://dl.acm.org/doi/pdf/10.1145/2641562 https://dl.acm.org/doi/pdf/10.1145/2641562
- rhindi 4y agoIt’s much much faster now, and performance is improving 10x every couple of year. With the current trend, FHE will be applicable to 80% of usecases by 2025
- nomnoms 4y agoPerformance is still problematic for many applications (particularly in ML). Our team has been working on making FHE more accessible to engineers via a compiler; we've found usability to be a much bigger obstacle than performance. You might be surprised to see how far performance has come! For (an admittedly small example of) matrix-vector multiplication, we can do key generation, encryption, computation, decryption, and compilation in less than 5 seconds on a MacBook [0]. [0]: https://playground.sunscreen.tech/ https://playground.sunscreen.tech/