Learning multisensory integration and coordinate transformation via density estimation.
Learning multisensory integration and coordinate transformation via density estimation.
复制标题
DOI：
10.1371/journal.pcbi.1003035
复制
发表时间：
2013-04
影响因子：
4.3
通讯作者：
Sabes PN
中科院分区：
文献类型：
--
作者：
Makin JG;Fellows MR;Sabes PN
大脑中的感觉处理包括三个关键操作：多感觉整合——将线索组合成对共同潜在刺激的单一估计；坐标变换——通过对干预变量（如凝视位置）的了解，刺激（如视网膜中心到身体中心）的参照系发生变化；以及先验信息的结合。统计上最优的感觉处理要求这些操作在刺激上保持正确的后验分布。这种最优性的要素已经在人类和其他动物的许多行为环境中得到了证明，这表明神经计算确实是最优的。感觉模式之间的关系既复杂又具有可塑性，这进一步表明这些计算是习得的——但如何习得呢？我们提供了一个原则性的答案，通过将这些映射的获取作为密度估计的一个案例，密度估计是机器学习和统计学中一个得到充分研究的问题，其中观测数据的分布是根据一组固定参数和一组潜在变量来建模的。在我们的案例中，观察到的数据是单感觉种群活动，固定参数是突触连接，潜在变量是多感觉种群活动。特别是，我们用生物学上合理的对比发散规则训练了一个受限的玻尔兹曼机，以学习一系列以前没有在单一方法下展示的神经计算：最优积分；先验编码；线索的层次整合；学习的时候不去整合；还有坐标变换。该模型对多感官表征的本质做出了可测试的预测。在生命的最初几年里，人类（和其他动物）似乎学会了如何将来自多种感觉模式的信号组合在一起：何时将它们“整合”成一个单一的感知，比如关于身体的视觉和本体感受信息；何时不整合它们（例如，当寻找其他地方时）；它们在更长的时间尺度上是如何变化的（例如，我的手在物理空间中的位置）；以及更复杂的操作，如从物体的视觉感知位置减去凝视角度，以计算该物体相对于头部的位置。，“坐标变换”。学习整合哪些感官信号，或者以其他方式操作哪些感官信号，似乎不需要额外的监督信号；我们学会这样做，更确切地说，是基于感官信号本身的结构。我们提出了一个生物学上合理的人工神经网络，它以这种方式学习上述所有内容，但通过训练它进行更一般的统计任务：“密度估计”——本质上，学习能够复制训练它的数据。这也将坐标转换和多感觉整合与其他皮层操作联系起来，特别是在早期感觉区域，这些操作已经被建模为密度估计器。
Sensory processing in the brain includes three key operations: multisensory integration—the task of combining cues into a single estimate of a common underlying stimulus; coordinate transformations—the change of reference frame for a stimulus (e.g., retinotopic to body-centered) effected through knowledge about an intervening variable (e.g., gaze position); and the incorporation of prior information. Statistically optimal sensory processing requires that each of these operations maintains the correct posterior distribution over the stimulus. Elements of this optimality have been demonstrated in many behavioral contexts in humans and other animals, suggesting that the neural computations are indeed optimal. That the relationships between sensory modalities are complex and plastic further suggests that these computations are learned—but how? We provide a principled answer, by treating the acquisition of these mappings as a case of density estimation, a well-studied problem in machine learning and statistics, in which the distribution of observed data is modeled in terms of a set of fixed parameters and a set of latent variables. In our case, the observed data are unisensory-population activities, the fixed parameters are synaptic connections, and the latent variables are multisensory-population activities. In particular, we train a restricted Boltzmann machine with the biologically plausible contrastive-divergence rule to learn a range of neural computations not previously demonstrated under a single approach: optimal integration; encoding of priors; hierarchical integration of cues; learning when not to integrate; and coordinate transformation. The model makes testable predictions about the nature of multisensory representations. Over the first few years of their lives, humans (and other animals) appear to learn how to combine signals from multiple sense modalities: when to “integrate” them into a single percept, as with visual and proprioceptive information about one's body; when not to integrate them (e.g., when looking somewhere else); how they vary over longer time scales (e.g., where in physical space my hand tends to be); as well as more complicated manipulations, like subtracting gaze angle from the visually-perceived position of an object to compute the position of that object with respect to the head—i.e., “coordinate transformation.” Learning which sensory signals to integrate, or which to manipulate in other ways, does not appear to require an additional supervisory signal; we learn to do so, rather, based on structure in the sensory signals themselves. We present a biologically plausible artificial neural network that learns all of the above in just this way, but by training it for a much more general statistical task: “density estimation”—essentially, learning to be able to reproduce the data on which it was trained. This also links coordinate transformation and multisensory integration to other cortical operations, especially in early sensory areas, that have have been modeled as density estimators.
登录
查看更多内容
影响因子：
5.4
作者：
ATTNEAVE, F
通讯作者：
ATTNEAVE, F
影响因子：
4.3
作者：
Eichhorn J;Sinz F;Bethge M
通讯作者：
Bethge M
影响因子：
25
作者：
Deneve, S;Latham, PE;Pouget, A
通讯作者：
Pouget, A
DOI：
10.1037/h0040546
发表时间：
1963-01-01
期刊：
JOURNAL OF COMPARATIVE AND PHYSIOLOGICAL PSYCHOLOGY
影响因子：
--
作者：
HELD, R;HEIN, A
通讯作者：
HEIN, A
DOI：
10.1073/pnas.0913209107
发表时间：
2010-04-27
影响因子：
11.1
作者：
Chang, Steve W. C.;Snyder, Lawrence H.
通讯作者：
Snyder, Lawrence H.