14 ms·
Yes and every halving of that 1% error is more than twice as hard to do with neural net training. At some point more principled control algorithms are needed.
by samegene321 2y ago
Yes and every halving of that 1% error is more than twice as hard to do with neural net training. At some point more principled control algorithms are needed.