Mapping biological neuronal superpower onto ( for now ) physical chips means completely ignoring accepted silicon architectures and Memristor innovations are likely to make it happen. But how?
Even with the juggernaut industrial shift to GPU based architectures and that has allowed Ai to ( almost ) literally pop into reality, it's power consumption that is always the killer issue even more so than error correcting and which for Memristors ( that have incredibly low energy consumption ) one of the most important factors in the architectural shift coming means we are edging closer to ( possibly ) the greatest change possible in computation: non-biological ( for now ) neuronal computing. In the past ( and even with original silicon computing and onto quantum error correction system making breakthroughs today ) as we get closer with neuromorphic computing ( and a key component of that, is Memristor technology ) with one of it's the last stages to adoption being that error correcting being locked down.
In a most recent breakthrough and as detailed in ...Computing Platform[s] Based On A Selector-less Memristor Array... the jump to domain non-specific architecture possibilities can be seen and with them any number of platforms are possible that may vastly rival current Ai based systems. These new architectures will allow for new and radical, incredibly low power consumption never been seen before ways to tackle some of the largest and most difficult design problems but more, engineering specific situations that require infinitely powerful computation with little or no power consumption and as detailed more in Mem-Systems And Neuromorphic Applications
Where the true breakthrough lay, will be in the shifts ( both in neuromorphic programing languages and applications ) and via the capabilities for Neuromorphic System with Online Learning at the forefront of usage most likely as the capability for such systems to rapidly and with very little power be able to adapt and morph into situational functionality will allow businesses, engineering and design capabilities in computation of achieve never before levels of sporadic and generative functionality and which is right around the corner.
#iGNITIATE #Design #DesignThinking #DesignInnovation #IndustrialDesign #iGNITEconvergence #iGNITEprogram #DesignLeadership #LawrenceLivermoreNationalLabs #NSF #USNavy #EcoleDesPonts #Topiade #LouisVuitton #WorldRetailCongress #REUTPALA #WorldRetailCongress #OM #Fujitsu #Sharing #Swarovski #321-Contact #Bausch&Lomb #M.ONDE #SunStar