School of Psychology and Neuroscience, University of Glasgow, Glasgow G12 8QB, UK.
School of Psychology and Neuroscience, University of Glasgow, Glasgow G12 8QB, UK.
Trends Cogn Sci. 2022 Dec;26(12):1090-1102. doi: 10.1016/j.tics.2022.09.003. Epub 2022 Oct 7.
Deep neural networks (DNNs) have become powerful and increasingly ubiquitous tools to model human cognition, and often produce similar behaviors. For example, with their hierarchical, brain-inspired organization of computations, DNNs apparently categorize real-world images in the same way as humans do. Does this imply that their categorization algorithms are also similar? We have framed the question with three embedded degrees that progressively constrain algorithmic similarity evaluations: equivalence of (i) behavioral/brain responses, which is current practice, (ii) the stimulus features that are processed to produce these outcomes, which is more constraining, and (iii) the algorithms that process these shared features, the ultimate goal. To improve DNNs as models of cognition, we develop for each degree an increasingly constrained benchmark that specifies the epistemological conditions for the considered equivalence.
深度神经网络 (DNN) 已经成为建模人类认知的强大且日益普及的工具，并且经常产生类似的行为。例如，通过其分层的、受大脑启发的计算组织，DNN 显然以与人类相同的方式对现实世界的图像进行分类。这是否意味着它们的分类算法也相似？我们通过三个嵌入的程度来提出这个问题，这些程度逐步限制算法相似性评估：（i）行为/大脑反应的等价性，这是当前的实践，（ii）为产生这些结果而处理的刺激特征，这更具约束性，以及（iii）处理这些共享特征的算法，这是最终目标。为了提高 DNN 作为认知模型的性能，我们为每个程度开发了一个越来越受约束的基准，该基准指定了所考虑等价的认识论条件。
