#science #genetics #nature #molecularbiology #innovation #incrediblecreatures #latestresearchesandnews #groundbreakingnews #latestreseachinscience #recentscientifictechnologydevelopment #scientificreasons #scientificwonders #sciencefacts #scienceandtechnologynews #cuttingedgescience #techinnovation #futuretech #scienceexploration #discoveriesunveiled #BioTechRevolution #naturewonders #InnovativeResearch #mindblowingscience #NextGenTechnology #scientificadvancements #BiomolecularWonders #exploretheunknown #ScienceNewsUpdate #FascinatingResearch #BiotechMarvels #MolecularDiscoveries #sciencegeek #techmarvels #sciencegeeks #NeuralCircuit #AI #NeuromorphicComputing #BrainInspiredTechnology #Dendristor #EnergyEfficientAI #TsinghuaUniversity #NatureElectronics #InnovativeAI #neuralnetworks

Researchers at Tsinghua University have developed a new neuromorphic computational architecture inspired by the brain's structure, particularly dendrites and synapses, using multi-gate silicon nanowire transistors with ion-doped sol-gel films. Published in Nature Electronics, this "dendristor" device mimics dendritic computations, allowing more efficient AI processing. The device leverages the movement of ions to modulate the transistor's current, achieving nonlinear dendritic integration and direction selectivity. This innovation, mirroring the sparse connectivity of brain neurons, promises significant energy efficiency and opens pathways for advanced neuromorphic systems and AI tools.

[ Ссылка ]

свернуть