推断鱼皮层中的种群动态
Ganga Meghanath1, Bryan Jimenez1, Joseph G Makin1
1Elmore Family School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN, United States of America.
Journal of neural engineering
|October 24, 2023
概括
循环神经网络 (RNN) 擅长模拟神经群体动态,在基准任务中表现优于其他模型. 一种结合RNN与自我注意力的新型混合架构进一步提高了预测神经活动的性能.
科学领域:
- 计算神经科学是一种神经科学.
- 机器学习 机器学习
- 神经工程 神经工程是神经工程.
背景情况:
- 多个单元皮质记录的数量增加,引发了对神经群体动态的兴趣.
- 模拟神经群体动态需要推断未被观察到的神经元活动并预测未来的状态.
- 通过在运动任务中使用子皮质神经记录来建立一个基准.
研究的目的:
- 评估一般性循环神经网络 (RNNs) 与神经群体动态的专门模型相比.
- 引入一种新的混合架构 (TERN),将RNN与自我注意力结合起来,以提高性能.
- 在神经潜伏基准上建立最先进的性能.
主要方法:
- 利用反复神经网络 (RNN) 与掩盖用于歧视性学习.
- 开发了一种新的混合架构 (TERN),用自我注意机制来增强RNN.
- 在皮神经记录的四个基准数据集上评估模型.
主要成果:
- 用掩盖训练的RNN在所有四个基准数据集上都超过了以前发布的所有模型.
- 混合架构TERN显示了进一步的性能改进.
- 纯变压器模型没有达到可比的性能水平.
结论:
- 在RNNs中固有的自回归偏差对于在建模神经群体动态中实现峰值性能至关重要.
- 这项研究在神经潜伏基准上建立了新的最先进的技术.
- 作者建议增加基准,以包括有利于生成模型的评估.
相关概念视频
Fixed Action Patterns
A fixed action pattern (FAP) is a specific, hard-wired sequence of behaviors that occurs in response to an external stimulus, called a sign stimulus. The behavior is “fixed” because it is essentially unchangeable—proceeding similarly across individuals of a species every time it occurs.
Speciation Rates
Speciation can proceed at markedly different rates, and evolutionary biologists commonly describe these differences through the models of gradualism and punctuated equilibrium. Both patterns explain how new species arise, but they differ in the tempo and continuity of evolutionary change. In both cases, evolutionary change arises from heritable variation within populations, with natural selection often shaping traits that improve survival and reproduction under specific environmental conditions.
Types of Selection
Natural selection influences the frequencies of particular alleles and phenotypes within populations in several different ways. Primarily, natural selection can be directional, stabilizing, or disruptive. Directional selection favors one extreme trait and shifts the population towards that phenotype while selecting against individuals displaying alternate traits. Stabilizing selection favors an intermediate trait with a narrow range of variation. Deviation from the optimal phenotype towards an...


