Moire synapse transistor developed using Moiré pattern
Inspired by the human brain, researchers have developed a new synaptic transistor that can simultaneously process and store information like the human brain, allowing for higher-level thinking.
Traditional computing vs brain
For decades, the paradigm in electronics has been to use transistors to make everything. Scientists have made significant progress simply by packing more and more transistors into integrated circuits.But this success isat the expense of high power consumption. In the current era of big data, the power grid is coming under pressure from digital computing. Recent advances in artificial intelligence (AI) and machine learning have made scientists increasingly eager to develop computing devices that operate more like the human brain.
The human brain is fundamentally different from a digital computer. In a traditional digital computer,The data processing unit and storage unit are independent, data consumes a lot of energy as it moves back and forth between the microprocessor and memory. This high power consumption problem is particularly serious when the computer has to perform multiple tasks at the same time. And in my mind,Memory and information processing are co-located and fully integratedthus having a higher order of magnitude energy efficiency.
In many past studies, some teams developed computing devices similar to the human brain, but thoseTransistors cannot operate in non-cold environments.
moiré synapse transistor
In the new study, the researchers decided to use moiré patterns to develop so-called synaptic transistors.
A moiré pattern is a geometric pattern that appears when two patterns are superimposed on each other. When two-dimensional materials are stacked, new properties emerge that are not present in a single-layer structure. When these hierarchical structures are twisted to form moiré patterns, it is possible to harvest unprecedented electronic properties.
In this new synaptic transistor, the researchers combined two different types of atomically thin materials:Double layer graphene and hexagonal boron nitride. Graphene and hexagonal boron nitride are structurally very similar, but they are also different enough to produce an unusually strong Moiré effect.
When the researchers stacked them on top of each other and twisted one layer of material relative to the other, the materials formed moiré patterns. By using this twist, researchers can achieve different electronic properties in each layer of the material, even if the layers are only separated by atomic-scale distances.
When twisting becomes a new design parameter, scientists can gain access to an enormous number of arrangements. Researchers found thatBy choosing the right twist, the newly developed synaptic transistor can remain stable at room temperature, operate quickly, consume very little energy, and retain stored information even when the power is turned off..
higher level thinking
If AIs are meant to exist to imitate human thinking, one of the lowest-level tasks they need to accomplish is classifying data. The researchers' goal is to push artificial intelligence technology toward higher-level thinking.
Real-world situations are often more complex than current artificial intelligence algorithms can handle. To test the new transistor, the team trained it to recognize similar but not identical patterns, testing its functionality in more complex situations.
First, they presented a pattern "000" to the device. They then asked the AI to identify similar patterns like "111" or "101". If the training content is to let it detect 000, then when it detects 111 and 101, it can recognize that 111 is more like 000 than 101. Although 000 and 111 are not exactly the same, they are both consecutive three digits.Recognizing this similarity is a higher-level form of cognition, which is calledassociative learning.
Experimental results showed that the new synaptic transistor successfully recognized similar patterns, demonstrating associative memory. This shows that the transistor has gone beyond simple machine learning tasks and can perform associative learning while classifying data.This means that, like the human brain, it implementsConcurrent memory and information processing capabilitiesmore closely imitates the human brain.
Current AI is prone to confusion, which can cause major problems in some cases. But this new type of synaptic transistor takes machine learning and artificial intelligence another step forward. It shows that even if the input we give a transistor is imperfect, it can still recognize and respond correctly.
Review Editor: Liu Qing
#Moire #synapse #transistor #developed #Moiré #pattern
- High-speed current impact test system facilitates efficient testing of new energy vehicle wiring harnesses
- Area array CCD structure diagram and working process analysis
- Input-output relationship and circuit application diagram of voltage follower
- How to keep driving heavy loads when the voltage drops?
- Novel theory-based evaluation gives a clearer picture of fusion in the sun
- What are the disadvantages of IC temperature sensors?
- Control transformer overcurrent protection, grounding and applications
- What is the difference between chip packaging and SMD?
- Design and application of dry multi-channel priority amplifier
- Can igbt directly replace thyristor? What will be the impact of IGBT directly replacing thyristor?