Optimal Encoding in Stochastic Latent-Variable Models.

Optimal Encoding in Stochastic Latent-Variable Models.
复制标题

DOI:
10.3390/e22070714
复制
发表时间:
2020-06-28
期刊:
Entropy (Basel, Switzerland)
影响因子:
--
通讯作者:
Hennig MH
Hennig MH
中科院分区:
其他
文献类型:
--
作者:
Rule ME;Sorbaro M;Hennig MH

文献摘要

参考文献

相似文献

在这项工作中,我们探讨了在嘈杂的尖峰网络的感觉编码的统计模型学习的编码策略。神经系统中的感觉通信的早期阶段可以被视为信息理论意义上的编码通道。然而,神经群体面临的限制,通常不考虑在通信理论。使用受限玻尔兹曼机作为感官编码的模型,我们发现,具有足够容量的网络学习平衡精度和噪声鲁棒性,以便自适应地传达具有不同信息内容的刺激。在感觉系统中观察到的镜像变异抑制,信息刺激以高精度编码,代价是对频繁的、因此信息较少的刺激的更多可变反应。奇怪的是,我们还发现神经群体代码中的统计临界性出现在输入统计数据被很好地捕获的模型大小上。这些现象有明确的热力学解释,我们讨论了他们的连接到流行的神经种群的编码和统计临界理论。
In this work we explore encoding strategies learned by statistical models of sensory coding in noisy spiking networks. Early stages of sensory communication in neural systems can be viewed as encoding channels in the information-theoretic sense. However, neural populations face constraints not commonly considered in communications theory. Using restricted Boltzmann machines as a model of sensory encoding, we find that networks with sufficient capacity learn to balance precision and noise-robustness in order to adaptively communicate stimuli with varying information content. Mirroring variability suppression observed in sensory systems, informative stimuli are encoded with high precision, at the cost of more variable responses to frequent, hence less informative stimuli. Curiously, we also find that statistical criticality in the neural population code emerges at model sizes where the input statistics are well captured. These phenomena have well-defined thermodynamic interpretations, and we discuss their connection to prevailing theories of coding and statistical criticality in neural populations.
DOI: 10.1103/physrevlett.97.118102
发表时间: 2006-09-15
影响因子: 8.6
作者:
Bedard, C.;Kroger, H.;Destexhe, A.
通讯作者: Destexhe, A.
DOI: 10.1068/p010371
发表时间: 1972-01-01
期刊: Perception
影响因子: 1.7
作者:
Barlow, H B
通讯作者: Barlow, H B
DOI: 10.1561/2200000006
发表时间: 2009-01-01
影响因子: 32.8
作者:
Bengio, Yoshua
通讯作者: Bengio, Yoshua
DOI: 10.1007/s10955-017-1770-6
发表时间: 2017-05
影响因子: 1.6
作者:
Bradde S;Bialek W
通讯作者: Bialek W
DOI: 10.1088/1742-5468/ab16c8
发表时间: 2019-06-01
影响因子: 2.4
作者:
Cubero, Ryan John;Jo, Junghyo;Song, Juyong
通讯作者: Song, Juyong