设计尖端神经网络,以实现强大且可重新配置的计算
Georg Börner1, Fabio Schittler Neves1,2, Marc Timme1,3
1Chair for Network Dynamics, Institute for Theoretical Physics and Center for Advancing Electronics Dresden (CFAED), TUD Dresden University of Technology, 01062 Dresden, Germany.
本研究提出了一种分析方法,用于调整尖端神经网络参数,以确保可靠的计算. 它侧重于k-winners-takes-all (k-WTA) 任务,对于弹性计算系统至关重要.
科学领域:
- 计算神经科学是一种计算神经科学.
- 人工智能的人工智能是人工智能.
- 复杂的系统复杂的系统.
背景情况:
- 尖端神经网络 (SNN) 提供了弹性模拟计算.
- 对称连接的抑制性SNN可以进行稳健的计算,即使有神经元损失.
- 适应网络参数以应对中断仍然是一个挑战.
研究的目的:
- 开发一种分析方法来导出SNN中的网络参数.
- 研究SNN中k-winners-takes-all (k-WTA) 计算的动力学.
- 为了使SNN计算系统能够设计抗中断的设计.
主要方法:
- 在SNN中对k-winners-takes-all (k-WTA) 计算进行分析.
- 网络内的不同动态模式的特征.
- 分析表达式的推导,用于k-赢家状态之间的过渡.
主要成果:
- 在k-WTA计算中识别不同的动态模式.
- 基于输入和网络属性的参数转换所提供的分析表达式.
- 展示如何调整参数以提供强大的k-WTA功能.
结论:
- 这项研究提供了对k-WTA计算动态的分析见解.
- 提供了一种设计具有抗中断活力的SNN动态的方法.
- 增强对参数适应的理解,以实现可靠的SNN计算.
更多相关视频
07:34A Simple Stimulatory Device for Evoking Point-like Tactile Stimuli: A Searchlight for LFP to Spike Transitions
Published on: March 25, 2014
10:32Design, Surface Treatment, Cellular Plating, and Culturing of Modular Neuronal Networks Composed of Functionally Inter-connected Circuits
Published on: April 15, 2015
相关概念视频
Neuroplasticity
Neural Circuits
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...
The Role of Ion Channels in Neuronal Computation
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....
Ampere-Maxwell's Law: Problem-Solving
To solve the problem, we can use the equations from the analysis of an RC circuit and Maxwell's version of Ampère's law.
For the first part of...
