Related Experiment Video
Updated: Aug 25, 2025
07:38
Author Spotlight: Modular Neuronal Networks for Analyzing Brain Functions
Published on: June 7, 2024
1.7K
Two-dimensional materials-based probabilistic synapses and reconfigurable neurons for measuring inference uncertainty
Amritanand Sebastian1, Rahul Pendurthi2, Azimkhan Kozhakhmetov3
1Deparment of Engineering Science and Mechanics, Penn State University, University Park, PA, 16802, USA. amritsebastian@gmail.com.
Nature Communications
|October 17, 2022
Summary
Researchers developed novel memristors from 2D materials to create Bayesian neural networks (BNNs). These BNNs can emulate probabilistic synapses and neurons, improving prediction reliability for critical applications.
Area of Science:
- Materials Science
- Computer Science
- Artificial Intelligence
Background:
- Traditional artificial neural networks lack uncertainty quantification, leading to high-confidence incorrect predictions.
- Bayesian neural networks (BNNs) address this by representing weights as probability distributions, inherently including uncertainty.
- Mission-critical applications require reliable predictions with confidence estimation.
Purpose of the Study:
- To introduce three-terminal memtransistors based on two-dimensional (2D) materials.
- To emulate probabilistic synapses and reconfigurable neurons for Bayesian neural networks.
- To realize a BNN accelerator for data classification tasks.
Main Methods:
- Exploiting cycle-to-cycle variation in 2D memtransistor programming for Gaussian random number generator-based synapses.
- Utilizing 2D memtransistor integrated circuits for neurons with hyperbolic tangent and sigmoid activation functions.
- Integrating memtransistor-based synapses and neurons in a crossbar array architecture.
Main Results:
- Demonstrated 2D memtransistors capable of emulating probabilistic synapses and neurons.
- Successfully implemented Gaussian random number generation and specific activation functions using memtransistors.
- Constructed a BNN accelerator using memtransistor synapses and neurons for a data classification task.
Conclusions:
- 2D memtransistors offer a promising platform for creating hardware accelerators for Bayesian neural networks.
- The developed memtransistor-based BNNs can perform data classification tasks while incorporating uncertainty estimation.
- This work paves the way for more reliable and robust AI systems in critical applications.
Related Concept Videos
01:25Neural Circuits
1.5K
Neural circuits and neuronal pools are two of the main structures found in the nervous system. Neural circuits are networks of neurons that work together to carry out a specific task or process. They consist of interconnected neurons and glial cells, which provide structural and metabolic support.
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
1.5K
00:59Propagation of Uncertainty from Random Error
936
An experiment often consists of more than a single step. In this case, measurements at each step give rise to uncertainty. Because the measurements occur in successive steps, the uncertainty in one step necessarily contributes to that in the subsequent step. As we perform statistical analysis on these types of experiments, we must learn to account for the propagation of uncertainty from one step to the next. The propagation of uncertainty depends on the type of arithmetic operation performed on...
936
01:01Neuroplasticity
693
Neuroplasticity reflects the brain's remarkable capacity to adapt and evolve, responding dynamically to learning, experiences, or injury by reorganizing its neural circuitry. This reorganization involves creating new neural connections and refining old ones through a series of biological processes that contribute to the brain's lifelong development and adaptability.
693
01:10Propagation of Uncertainty from Systematic Error
750
The atomic mass of an element varies due to the relative ratio of its isotopes. A sample's relative proportion of oxygen isotopes influences its average atomic mass. For instance, if we were to measure the atomic mass of oxygen from a sample, the mass would be a weighted average of the isotopic masses of oxygen in that sample. Since a single sample is not likely to perfectly reflect the true atomic mass of oxygen for all the molecules of oxygen on Earth, the mass we obtain from this...
750
01:28Electrical Synapses
8.6K
Electrical synapses found in all nervous systems play important and unique roles. In these synapses, the presynaptic and postsynaptic membranes are very close together (3.5 nm) and are actually physically connected by channel proteins forming gap junctions.
Gap junctions allow the current to pass directly from one cell to the next. In contrast, in the chemical synapse, the neurotransmitters carry the information through the synaptic cleft from one neuron to the next. They consist of two...
Gap junctions allow the current to pass directly from one cell to the next. In contrast, in the chemical synapse, the neurotransmitters carry the information through the synaptic cleft from one neuron to the next. They consist of two...
8.6K
00:59Uncertainty: Overview
785
In analytical chemistry, we often perform repetitive measurements to detect and minimize inaccuracies caused by both determinate and indeterminate errors. Despite the cares we take, the presence of random errors means that repeated measurements almost never have exactly the same magnitude. The collective difference between these measurements - observed values - and the estimated or expected value is called uncertainty. Uncertainty is conventionally written after the estimated or expected value.
785