Incorporating intrinsic suppression in deep neural networks captures dynamics of adaptation in neurophysiology and perception.
Incorporating intrinsic suppression in deep neural networks captures dynamics of adaptation in neurophysiology and perception.
复制标题
DOI：
10.1126/sciadv.abd4205
复制
发表时间：
2020-10
期刊：
影响因子：
13.6
通讯作者：
Kreiman G
中科院分区：
文献类型：
--
作者：
Vinken K;Boix X;Kreiman G
在人工神经网络中添加细胞适应机制解释了生物视觉的时间背景调制。适应是感觉系统的基本属性，可以改变最近信息背景下的主观体验。适应被认为是由循环回路机制或神经元内在抑制引起的。然而，目前尚不清楚内在抑制本身是否可以解释除了反应减少之外的影响。在这里，我们测试了这样的假设：复杂的适应现象可以通过视觉处理的前馈模型从内在抑制级联中出现。具有内在抑制功能的深度卷积神经网络捕获了适应的神经特征，包括新颖性检测、增强和调整曲线移动，同时产生与人类感知一致的后效。当在重复输入影响识别性能的任务中训练适应性时，内在机制比循环神经网络具有更好的泛化能力。我们的结果表明，内在抑制的前馈传播改变了网络的功能状态，再现了适应的关键神经生理学和感知特性。
Adding a cellular adaptation mechanism to artificial neural networks explains temporal context modulation of biological vision. Adaptation is a fundamental property of sensory systems that can change subjective experiences in the context of recent information. Adaptation has been postulated to arise from recurrent circuit mechanisms or as a consequence of neuronally intrinsic suppression. However, it is unclear whether intrinsic suppression by itself can account for effects beyond reduced responses. Here, we test the hypothesis that complex adaptation phenomena can emerge from intrinsic suppression cascading through a feedforward model of visual processing. A deep convolutional neural network with intrinsic suppression captured neural signatures of adaptation including novelty detection, enhancement, and tuning curve shifts, while producing aftereffects consistent with human perception. When adaptation was trained in a task where repeated input affects recognition performance, an intrinsic mechanism generalized better than a recurrent neural network. Our results demonstrate that feedforward propagation of intrinsic suppression changes the functional state of the network, reproducing key neurophysiological and perceptual properties of adaptation.
登录
查看更多内容
影响因子：
25
作者：
Kar K;Kubilius J;Schmidt K;Issa EB;DiCarlo JJ
通讯作者：
DiCarlo JJ
影响因子：
3.7
作者：
Jeyabalaratnam J;Bharmauria V;Bachatene L;Cattan S;Angers A;Molotchnikoff S
通讯作者：
Molotchnikoff S
影响因子：
4.3
作者：
Kalfas I;Vinken K;Vogels R
通讯作者：
Vogels R
影响因子：
9.2
作者：
Bell, Andrew H.;Summerfield, Christopher;Morin, Elyse L.;Malecek, Nicholas J.;Ungerleider, Leslie G.
通讯作者：
Ungerleider, Leslie G.
影响因子：
4.3
作者：
Khaligh-Razavi SM;Kriegeskorte N
通讯作者：
Kriegeskorte N