Working on parameterizing a search space that includes more than the local group size. The end dream is some ML guided search to optimize the kernels :)
OK generally I think you're doing exactly what I believe ML is lacking right now. Another huge opportunity is instead of taking the average neural network and designing accelerators for it, designing hardware-friendly networks that run well on a sane accelerator that was designed to work with only these specialised networks (that doesn't need 80% chip area for on-chip memory for example). These might end up being completely different networks to what researchers use today. I work in this area and I think it's also possible to use the loss function to optimise the network for a specific HW.
Working on parameterizing a search space that includes more than the local group size. The end dream is some ML guided search to optimize the kernels :)