Ask HN: Any research labs/companies using neuromorphic chips in production?
There's been a lot of progress on neuromorphic computing and some companies are providing solid solutions (e.g., Intel Loihi). In theory NM should improve speed of inference and training and cut down the costs of running neural nets. Does anyone on this community have experience using NM chips in production environments for real-world use cases or for research? If you do, what is your take? If you don't, what's the main barrier?
Algorithms to successfully perform any practical tasks with spiking nets do not exist yet. We have very little clue how a brain uses spikes. As a general rule, you don't want to build hardware to accelerate something that does not work in software.
Source: I worked in a lab where we have built memristor based chips to run spiking nets for image classification and other tasks (my role was developing training algorithms).