Hi FF,
So many fundamental problems to overcome:
Indeed, researchers have yet to figure out simple ways to get neuromorphic systems to run the huge variety of deep-learning networks that have been developed on conventional chips. Brian Van Essen, at the Center for Applied Scientific Computing at the Lawrence Livermore National Laboratory, said his group has been able to get neural networks to run on TrueNorth but that the task of picking the right network and then successfully porting it over remains “a challenge." Other researchers say the most advanced deep-learning systems require more neurons, with more possible interconnections, than current neuromorphic technology can offer
AKIDA: 1-shot on-chip learning
By contrast, there is hardly a Webcompany or university computer department on the planet that isn't doing something with deep learning on conventional chips. As a result, those conventional architectures have a robust suite of development tools, along with legions of engineers trained in their use— typical advantages of an incumbent technology with a large installed base. Getting the deep-learning community to switch to a new and unfamiliar way of doing things will prove extremely difficult unless neuromorphics can offer an unmistakable performance and power advantage.
AKIDA: Meta TF - CNN-to-SNN
Some companies on the vanilla side of this argument deny that neuromorphic systems have an edge in power efficiency. William J. Dally, a Stanford electrical engineering professor and chief scientist at Nvidia, said that the demonstrations performed with TrueNorth used a very early version of deep learning, one with much less accuracy than is possible with more recent systems. When accuracy is taken into account, he said, any energy advantage of neuromorphics disappears.“People who do conventional neural networks get results and win the competitions," Dally said. “The neuromorphic approaches are interesting scientifically, but they are nowhere close on accuracy."
AKIDA: 1-bit to 4-bit weights and activations
The genius of PvdM's concept is the realization that, as opposed to mathematical calculations, sensor outputs corresponding to humans senses do not require mathematical accuracy of 8-bit or higher numbers, and, or many applications, 1-bit weights and actuations were sufficient. The expertise to design a circuit adapted to implement an array of such spike detectors/decoders have left the competition bobbing about in our wake.
(20min delay)