What AI, Neuroscience, and Cognitive Science Can Learn from Each Other: An Embedded Perspective

What AI, Neuroscience, and Cognitive Science Can Learn from Each Other: An Embedded Perspective
复制标题

人工智能、神经科学和认知科学可以互相学习什么:嵌入式视角

DOI:
10.1007/s12559-023-10194-9
复制
发表时间:
2023
影响因子:
5.4
通讯作者:
Achler, Tsvi
Achler, Tsvi
中科院分区:
计算机科学2区
文献类型:
--
作者:
Achler, Tsvi

文献摘要

参考文献

相似文献

研究人工智能和神经科学领域的科学家可以互相学习很多东西,但不幸的是,自 20 世纪 50 年代左右以来,这种情况大多是片面的:神经科学家从人工智能中学习,但从其他方面学习的却很少。我认为这阻碍了大脑的理解和人工智能的进步。当前的人工智能(“神经网络”/深度学习算法)和大脑彼此非常不同。大脑似乎并不使用反向传播等试错型学习算法来修改权重,更重要的是,不需要进行试错实现所需的繁琐排练。当遇到信息时,大脑可以以模块化和真正的“一次性”方式学习信息,而人工智能却不能。有证据表明,大脑在识别过程中使用调节反馈来调节其输入,而不是反向传播和排练:将输出形成回输入——激活输出的输入相同。这是通过神经科学和认知心理学领域的证据观察到的,但在当前的算法中并不存在。因此,大脑提供了关于其底层算法的大量证据,虽然计算机科学工具和分析至关重要,但计算机科学指导的算法不应标准化为神经科学理论。
Scientists studying in the fields of AI and neuroscience can learn much from each other, but unfortunately, since about the 1950s, it has been mostly one-sided: neuroscientists have learned from AI, but less so the other way. I argue this is holding back both brain understanding and progress in AI. Current AI (“neural network”/deep learning algorithms) and the brain are very different from each other. The brain does not seem to use trial-and-error–type learning algorithms such as backpropagation to modify weights and more importantly does not require the cumbersome rehearsal needed for trial-and-error implementation. The brain can learn information in a modular and true “one-shot” fashion as the information is encountered while the AI cannot. Instead of backpropagation and rehearsal, there is evidence that the brain regulates its inputs during recognition using regulatory feedback: form the outputs back to inputs—the same inputs that activate the outputs. This is observed through evidence from the fields of neuroscience and cognitive psychology but is not present in current algorithms. Thus, the brain provides an abundance of evidence about its underlying algorithms and while computer science tools and analysis are essential, algorithms guided by computer science should not be standardized into neuroscience theories.
石井,N.;A.W.M.辛普森;C.C.阿什利:科学。
DOI: --
发表时间: --
期刊:
影响因子: --
作者:
通讯作者: --
DOI: 10.1007/s13218-021-00705-x
发表时间: 2021-02-26
影响因子: 2.9
作者:
Butz, Martin V.
通讯作者: Butz, Martin V.
心理回放可实现灵活的导航
DOI: --
发表时间: 2022
期刊: Nature
影响因子: 64.8
作者:
J. Epsztein
通讯作者: J. Epsztein
目光转移的预测编码模型和潜在的神经生理学
DOI: 10.1080/13506285.2017.1336141
发表时间: 2017
期刊: Visual Cognition
影响因子: 2
作者:
Michael W. Spratling
通讯作者: Michael W. Spratling
DOI: 10.1146/annurev.ne.18.030195.001205
发表时间: 1995
影响因子: 13.9
作者:
R. Desimone;J. Duncan
通讯作者: R. Desimone;J. Duncan