Neuromorphic semiconductors, the next generation: 'Design possible if human brain's principles are understood'

Harvard professor Ham Don-hee has highlighted the potential of neuromorphic semiconductors, a next-generation technology designed to mimic the human brain's operational principles for enhanced computing efficiency. Speaking at a special lecture hosted by the Choi Soon-hyun Foundation in Seoul, Ham explained that current computer architectures separate memory and processing functions, leading to inefficiencies like signal delays and high power consumption.

In contrast, the human brain performs both memory and processing simultaneously. Ham believes that by understanding these brain functions more precisely, scientists can design future computing paradigms. His research focuses on the convergence of neuroscience and semiconductor engineering to advance this field.

“

Computers separate memory and processing units, but the brain does both simultaneously. If we understand the brain's operating principles, we can design future computing paradigms.

Ham, a distinguished scholar in neuromorphic semiconductors, previously served as the vice president of Samsung's SAIT (formerly Samsung Advanced Institute of Technology). His work on this subject was published in the prestigious journal Nature Electronics in 2021. He emphasized that a deeper understanding of how the brain learns, remembers, and makes decisions is key to designing advanced neuromorphic chips.

“

If we can understand more precisely how the brain learns, remembers, and makes decisions, designing next-generation neuromorphic chips will also be possible.