WASHINGTON (Dispatches) -- Scientists have reported that mimicking sleep patterns of the human brain in artificial neural networks may help mitigate the threat of catastrophic forgetting in the latter.
Researchers at University of California San Diego School of Medicine say that brain is very busy when asleep, repeating what has been learned during the day. Also, that Sleep helps reorganize memories and presents them in the most efficient way.
They explain how sleep builds rational memory, the ability to remember arbitrary or indirect associations between objects, people or events, and protects against forgetting old memories.
The researchers used spiking neural networks that artificially mimic natural neural systems: Instead of information being communicated continuously, it is transmitted as discrete events (spikes) at certain time points.
They found that when the spiking networks were trained on a new task, but with occasional off-line periods that mimicked sleep, catastrophic forgetting was mitigated. Like the human brain, said the study authors, “sleep” for the networks allowed them to replay old memories without explicitly using old training data.