Deep Neural Networks as Scientific Models.

By Radoslaw Martin Cichy, Daniel Kaiser. 2019

406 citations recorded by the source corpus.

Artificial deep neural networks (DNNs) initially inspired by the brain enable computers to solve cognitive tasks at which humans excel. In the absence of explanations for such cognitive phenomena, in turn cognitive scientists have started using DNNs as models to investigate biological cognition and its neural basis, creating heated debate. Here, we reflect on the case from the perspective of philosophy of science. After putting DNNs as scientific models into context, we discuss how DNNs can fruitfully contribute to cognitive science. We claim that beyond their power to provide predictions and explanations of cognitive phenomena, DNNs have the potential to contribute to an often overlooked but ubiquitous and fundamental use of scientific models: exploration.

Subfields: Artificial Intelligence, Biophysics, Cognitive Science & Psychology, Philosophy

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Biophysics

In Cognitive Science & Psychology

In Philosophy

Citations in the corpus

Shortcut learning in deep neural networks (2020, background): “They certainly do have the potential to solve complex challenges and serve as scientific models for prediction, explanation and exploration [84]—however, we must not confuse performance on a dataset with the acquisition of an underlying ability.”
The Notorious Difficulty of Comparing Human and Machine Perception (2020, background): “From the perspective of cognitive science, Cichy and Kaiser (2019) stress that Deep Learning models can serve as scientiﬁc models that not only provide both helpful predictions and explanations but that can also be used for exploration.”
Synthetic-Neuroscore: Using a neuro-AI interface for evaluating generative adversarial networks (2019, background): “With the success achieved by deep neural networks (DNNs) in areas including computer vision and natural language processing, the operation and functionality of DNNs and its connection with the human brain has been extensively studied and investigated in the literature [22, 23, 24, 25, 26, 27, 18, 28”
Integrative Benchmarking to Advance Neurally Mechanistic Models of Human Intelligence. (2020, background): “Following the practical success of these models, they are now the subject of much theoretical work (Banburski et al., 2019; Golowich et al., 2018; Belkin et al., 2019; Casper et al., 2019).”
THINGS: A database of 1,854 object concepts and more than 26,000 naturalistic object images (2019, background): “Deep convolutional neural networks (CNNs) offer a computational model of object recognition and–due to their excellent performance–have become very popular models not only in the field of computer science, but also in psychology and neuroscience [34–36] (for review, see [37, 38]).”