arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1802.04762cs.CV
收藏

用于目标识别的深度预测编码网络

Deep Predictive Coding Network for Object Recognition

Purdue University(普渡大学)

Haiguang Wen, Kuan Han, Junxing Shi, Yizhen Zhang, Eugenio Culurciello, Zhongming Liu

摘要:

基于神经科学中的预测编码理论,我们设计了一个双向递归神经网络,即深度预测编码网络(PCN)。它有前馈、反馈和循环连接。来自较高层的反馈连接携带其较低层表示的预测;前馈连接将预测误差携带到其较高层。给定图像输入,PCN运行自底向上和自顶向下计算的递归循环,以更新其内部表示,并减少每层自底向上输入和自顶向下预测之间的差异。在多个循环的递归更新之后,该表示用于图像分类。使用基准数据(CIFAR-10/100,SVHN和MNIST),发现PCN总是优于其仅前馈的对应模型:一个没有任何循环动力学机制的模型。随着时间的推移,它的性能往往会随着更多的计算周期而改善。简而言之,PCN重用单个体系结构来递归地运行自底向上和自顶向下的流程。作为一个动态系统,PCN可以展开为一个前馈模型,随着时间的推移变得越来越深,同时改进它的表示,以实现更准确和确定的对象识别。

英文摘要:

Based on the predictive coding theory in neuroscience, we designed a bi-directional and recurrent neural net, namely deep predictive coding networks (PCN). It has feedforward, feedback, and recurrent connections. Feedback connections from a higher layer carry the prediction of its lower-layer representation; feedforward connections carry the prediction errors to its higher-layer. Given image input, PCN runs recursive cycles of bottom-up and top-down computation to update its internal representations and reduce the difference between bottom-up input and top-down prediction at every layer. After multiple cycles of recursive updating, the representation is used for image classification. With benchmark data (CIFAR-10/100, SVHN, and MNIST), PCN was found to always outperform its feedforward-only counterpart: a model without any mechanism for recurrent dynamics. Its performance tended to improve given more cycles of computation over time. In short, PCN reuses a single architecture to recursively run bottom-up and top-down processes. As a dynamical system, PCN can be unfolded to a feedforward model that becomes deeper and deeper over time, while refining it representation towards more accurate and definitive object recognition.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑