arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2011.08674cs.LGcs.AIcs.CV
收藏

深度神经网络的数值性研究

On Numerosity of Deep Neural Networks

Shanghai Jiao Tong University(上海交通大学);McMaster University(麦克马斯特大学)

Xi Zhang, Xiaolin Wu

摘要:

最近,有一种挑衅性的说法被发表,即数字感在一个只为视觉物体识别而训练的深度神经网络中自发地出现。如果这是真的,对机器学习和认知科学等领域都具有深远的意义。在本文中,我们证明上述主张是不幸的是不正确的。支持这种说法的统计分析是有缺陷的,因为与对象识别网络中大量的神经元相比,用于识别数字感知神经元的样本集太小了。通过这种有缺陷的分析,人们可能会错误地识别出任何随机初始化的深度神经网络中的数字感知神经元,而这些神经元根本没有经过训练。有了上面的批评,我们提出了一个问题,如果一个深度卷积神经网络经过精心的数字训练会怎么样?我们的调查结果喜忧参半。即使在接受了描绘数字的图像的训练后,深度学习方法仍然难以获得数字的抽象概念,这是一项学龄前儿童轻松完成的认知任务。但另一方面,我们确实发现了一些令人鼓舞的证据,表明深度神经网络对小数字的分布偏移比大数字更鲁棒。

英文摘要:

Recently, a provocative claim was published that number sense spontaneously emerges in a deep neural network trained merely for visual object recognition. This has, if true, far reaching significance to the fields of machine learning and cognitive science alike. In this paper, we prove the above claim to be unfortunately incorrect. The statistical analysis to support the claim is flawed in that the sample set used to identify number-aware neurons is too small, compared to the huge number of neurons in the object recognition network. By this flawed analysis one could mistakenly identify number-sensing neurons in any randomly initialized deep neural networks that are not trained at all. With the above critique we ask the question what if a deep convolutional neural network is carefully trained for numerosity? Our findings are mixed. Even after being trained with number-depicting images, the deep learning approach still has difficulties to acquire the abstract concept of numbers, a cognitive task that preschoolers perform with ease. But on the other hand, we do find some encouraging evidences suggesting that deep neural networks are more robust to distribution shift for small numbers than for large numbers.

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。

↑