In a joint article published on Intelligence of the Machine of NatureResearchers at Intel Labs and Cornell University have shown Intel & # 39; s neuromorphic chip ability, Lohihi, to read and appreciate harmful chemicals in the presence of significant noise and explosions.
Loihi studied each fragrance with one sample, without disturbing her memory of the fragrances she had previously studied. It showed high acceptable accuracy when compared to standard country methods, including a deep learning solution that requires three training samples per class to reach the same level of classification accuracy.
"We are developing a neural skill in Loihi to mimic what happens in your brain when you smell something. This work is an excellent example of modern research in the fields of neuroscience and artificial intelligence and demonstrates Lohihi's ability to provide valuable discovery skills that can benefit many industries." says Nabil Imam, senior research scientist at Intel & # 39; s Neuromorphic Computing Laboratory.
About the research
Using the neural algorithm found in the design and architecture of the artificial brain circuit, researchers at Intel & Cornell trained Intel & # 39; s Loihi neuromorphic research chip to study and detect the odors of 10 harmful chemicals. To do so, the team used a data set that included the work of 72 chemical sensors to respond to these odors and prepared a circuit diagram for the diagnosis of Loihi-related infections. The chip quickly studied the neural representation of each scent and identified each odor, albeit highly visible, suggesting a promising future for the combination of neuroscience and artificial intelligence.