A Double-Layer Multi-Resolution Classification Model for Decoding Spatiotemporal Patterns of Spikes With Small Sample Size.
A Double-Layer Multi-Resolution Classification Model for Decoding Spatiotemporal Patterns of Spikes With Small Sample Size.
复制标题
一种用于小样本尖峰脉冲时空模式解码的双层多分辨率分类模型。
DOI：
10.1162/neco_a_01459
复制
发表时间：
2021-12-15
影响因子：
2.9
通讯作者：
Song, Dong
中科院分区：
文献类型：
--
作者：
She, Xiwei;Berger, Theodore W.;Song, Dong
我们建立了一个双层多时间分辨率分类模型解码单次试验的时空模式的尖峰。该模型以发放活动为输入信号，以二元行为或认知变量为输出信号，用双层集成分类器表示输入-输出映射。在第一层中，为了解决由于小样本和输入信号的维数很高而引起的欠定问题，使用B样条函数展开和L1正则化逻辑分类器来降低维数并产生稀疏模型估计。广泛的时间分辨率的神经功能包括通过使用大量的分类器与不同数量的B样条节点。每个分类器作为一个基础学习器，将时空模式分类为具有单一时间分辨率的输出标签的概率。一个自举聚集策略被用来减少这些分类器的估计方差。在第二层中，另一个L1正则化逻辑分类器将第一层分类器的输出作为输入，以生成最终的输出预测。该分类器作为Meta学习器，融合多个时间分辨率，将尖峰的时空模式分类为二进制输出标签。我们测试这个解码模型与合成和实验数据记录从大鼠和人类受试者执行记忆依赖的行为任务。实验结果表明，该方法能有效避免过拟合，并在小样本情况下得到准确的输出标签预测。通过提取和利用锋电位模式的多分辨率时空特征，双层多分辨率分类器的分类性能始终优于最好的单层单分辨率分类器。
We build a double-layer multiple temporal-resolution classification model for decoding single-trial spatio-temporal patterns of spikes. The model takes spiking activities as input signals and binary behavioral or cognitive variables as output signals and represents the input-output mapping with a double-layer ensemble classifier. In the first layer, to solve the underdetermined problem caused by the small sample size and the very high dimensionality of input signals, B-spline functional expansion and L1-regularized logistic classifiers are used to reduce dimensionality and yield sparse model estimations. A wide range of temporal resolutions of neural features are included by using a large number of classifiers with different numbers of B-spline knots. Each classifier serves as a base learner to classify spatio-temporal patterns into the probability of the output label with a single temporal resolution. A bootstrap aggregating strategy is used to reduce estimation variances of these classifiers. In the second layer, another L1-regularized logistic classifier takes outputs of first-layer classifiers as inputs to generate the final output predictions. This classifier serves as a meta learner that fuses multiple temporal resolutions to classify spatio-temporal patterns of spikes into binary output labels. We test this decoding model with both synthetic and experimental data recorded from rats and human subjects performing memory-dependent behavioral tasks. Results show that this method can effectively avoid overfitting and yield accurate prediction of output labels with small sample size. The double-layer multi-resolution classifier consistently outperforms the best single-layer single-resolution classifier by extracting and utilizing multi-resolution spatio-temporal features of spike patterns in the classification.
登录
查看更多内容
影响因子：
3
作者：
Park K;Song S;Hong I;Song B;Kim J;Park S;Lee J;Song S;An B;Kim J;Lee CJ;Shin KS;Choi S;Lee S
通讯作者：
Lee S
影响因子：
1.8
作者：
Humayun, MS;Weiland, JD;de Juan, E
通讯作者：
de Juan, E
影响因子：
2.9
作者：
Brown, EN;Barbieri, R;Frank, LM
通讯作者：
Frank, LM
DOI：
10.1073/pnas.0400162101
发表时间：
2004-03-02
影响因子：
11.1
作者：
Hampson, RE;Pons, TP;Deadwyler, SA
通讯作者：
Deadwyler, SA
影响因子：
14.4
作者：
Diamantidis, NA;Karlis, D;Giakoumakis, EA
通讯作者：
Giakoumakis, EA