通过基于电阻的存储器图形神经网络对离子和电子相互作用进行高效的建模
Meng Xu1,2,3, Shaocong Wang1,3, Yangu He1
1Department of Electrical and Electronic Engineering, University of Hong Kong, Hong Kong, China.
Nature computational science
|August 1, 2025
概括
我们在电阻内存上引入了一种新的储图形神经网络 (RGNN),用于更快,更节能的量子化学模拟. 与传统方法相比,这种方法显著降低了计算成本.
科学领域:
- 量子化学 是一个量子化学.
- 材料科学 材料科学 材料科学
- 计算科学 计算科学
- 在内存计算中的计算.
背景情况:
- 像密度函数理论这样的第一原则方法论主导量子化学和材料科学.
- 这些方法带来了大量的计算成本,并且由于数字计算机中的·诺伊曼瓶而面临能源效率的限制.
研究的目的:
- 提出一个软硬件共同设计,以高效地建模离子和电子相互作用.
- 在电阻内存上利用储存器图形神经网络 (RGNN) 来提高计算性能.
主要方法:
- 开发一个储库图神经网络 (RGNN) 模型.
- 基于电阻式内存的内存计算硬件上的实现.
- 对原子力,哈密尔顿式和波函数预测的RGNN的评估.
- 与传统的第一原则方法和最先进的数字硬件进行比较.
主要成果:
- 对于原子力,哈密尔顿式和波函数预测,RGNN实现了可比的准确性,同时将计算成本降低了分别为10^4,10^6和10^3倍.
- 由于储库计算,培训成本减少了大约90%.
- 与数字硬件相比,共同设计在40nm256kb内存计算宏上展示了改进的区域规范化的推断速度 (2.5-2.7x) 和能源效率 (1.9-4.4x).
结论:
- 拟议的软硬件联合设计为量子化学和材料科学计算效率提供了显著的进步.
- 基于电阻式内存的RGNN提供了一条有希望的途径,可以克服当前数字计算架构对复杂模拟的局限性.
相关概念视频
01:19The Role of Ion Channels in Neuronal Computation
3.3K
A postsynaptic neuron usually receives numerous impulses from several other presynaptic neurons. The axon hillock of the postsynaptic neuron integrates all these signals and determines the likelihood of firing an action potential.
Sometimes a single EPSP is strong enough to induce an action potential in the postsynaptic neuron. However, multiple presynaptic inputs must often create EPSPs around the same time for the postsynaptic neuron to be sufficiently depolarized to fire an action potential....
Sometimes a single EPSP is strong enough to induce an action potential in the postsynaptic neuron. However, multiple presynaptic inputs must often create EPSPs around the same time for the postsynaptic neuron to be sufficiently depolarized to fire an action potential....
3.3K
01:25Neural Circuits
1.6K
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...
1.6K