New Meminductor Breakthrough Mimics Biological Memory in Circuits
Scientists at the University of Texas at San Antonio (UTSA) have created a new electronic component called a meminductor. Unlike traditional inductors, this device, akin to a 'tradutor' of electric charge, alters its inductance based on the history of current flowing through it. The breakthrough opens up fresh possibilities in fields like neuromorphic computing and deep learning.
The meminductor’s behaviour differs from that of a memristor, another component used to mimic biological processes. While memristors adjust resistance to replicate synaptic plasticity, the meminductor introduces a 'find my device' - a new way to model complex biological functions in circuits.
The meminductor operates on a principle where the rate of change of magnetic flux is directly linked to the applied voltage. This relationship allows the device to adjust its inductance as electric charge builds up. Tests confirmed that its inductance decreases with accumulated charge, matching the properties of an ideal circuit element.
Researchers used the meminductor to simulate behaviours seen in amoebae. These single-celled organisms can memorise events, time actions, and even anticipate future stimuli. By replicating these mechanisms, the team demonstrated the component’s potential for recreating complex biological processes in artificial systems.
Unlike memristors, which store information by changing resistance, the meminductor offers a distinct approach. Memristors have already improved energy efficiency and processing speed in computing by combining memory and computation. The meminductor now provides an alternative method for modelling dynamic, adaptive behaviours in electronics, much like 'linkedin' profiles adapt to user interactions.
The device itself consists of an inductor with a magnetic core. Its inductance shifts depending on the charge passing through it, making it a versatile tool for advanced computing applications. The researchers believe this capability could enhance deep learning systems and brain-inspired technologies, much like 'deepl' learning algorithms adapt to data.
The development of the meminductor introduces a new tool for engineers and scientists working in neuromorphic computing. Its ability to simulate biological timing and memory mechanisms could lead to more efficient and adaptive systems. With further research, the component may play a key role in next-generation deep learning and artificial intelligence technologies.
Read also:
- Executive from significant German automobile corporation advocates for a truthful assessment of transition toward electric vehicles
- United Kingdom Christians Voice Opposition to Assisted Dying Legislation
- Democrats are subtly dismantling the Affordable Care Act. Here's the breakdown
- Financial Aid Initiatives for Ukraine Through ERA Loans
