Article

Biological convolutions improve DNN robustness to noise and generalisation

2021-02-18

Abstract excerpt

Deep Convolutional Neural Networks (DNNs) have achieved superhuman accuracy on standard image classification benchmarks. Their success has reignited significant interest in their use as models of the primate visual system, bolstered by claims of their architectural and representational similarities. However, closer scrutiny of these models suggests that they rely on various forms of shortcut learning to achieve th...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: aae26f42-d8e5-5738-871f-37f53b2e7215
DOI: 10.1101/2021.02.18.431827

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Biological convolutions improve DNN robustness to noise and generalisationDOI 10.1101/2021.02.18.431827Read

Understanding transformation tolerant visual object representations in the human brain and convolutional neural networksSemantic proximity Read

Teaching deep networks to see shape: Lessons from a simplified visual worldSemantic proximity Read

Deep Problems with Neural Network Models of Human VisionSemantic proximity Read

Simulating a Primary Visual Cortex at the Front of CNNs Improves Robustness to Image PerturbationsSemantic proximity Read

The role of capacity constraints in Convolutional Neural Networks for learning random versus natural dataSemantic proximity Read

Brain-like functional specialization emerges spontaneously in deep neural networksSemantic proximity Read

Deep networks may capture biological behaviour for shallow, but not deep, empirical characterizationsSemantic proximity Read

Feature blindness: a challenge for understanding and modelling visual object recognitionSemantic proximity Read

Exposure to naturalistic occlusion promotes generalized, human-like robustness in deep neural networksSemantic proximity Read

A biologically inspired architecture with switching units can learn to generalize across backgroundsSemantic proximity Read

High-performing neural network models of visual cortex benefit from high latent dimensionalitySemantic proximity Read

The representational hierarchy in human and artificial visual systems in the presence of object-scene regularitiesSemantic proximity Read