Model metamers reveal divergent invariances between biological and artificial neural networks
Deep neural network models of sensory systems are often proposed to learn
representational transformations with invariances like those in the brain. To reveal these …
representational transformations with invariances like those in the brain. To reveal these …
Silences, spikes and bursts: Three‐part knot of the neural code
Z Friedenberger, E Harkin, K Tóth… - The Journal of …, 2023 - Wiley Online Library
When a neuron breaks silence, it can emit action potentials in a number of patterns. Some
responses are so sudden and intense that electrophysiologists felt the need to single them …
responses are so sudden and intense that electrophysiologists felt the need to single them …
Aligning model and macaque inferior temporal cortex representations improves model-to-human behavioral alignment and adversarial robustness
While some state-of-the-art artificial neural network systems in computer vision are strikingly
accurate models of the corresponding primate visual processing, there are still many …
accurate models of the corresponding primate visual processing, there are still many …
Lcanets: Lateral competition improves robustness against corruption and attack
Abstract Although Convolutional Neural Networks (CNNs) achieve high accuracy on image
recognition tasks, they lack robustness against realistic corruptions and fail catastrophically …
recognition tasks, they lack robustness against realistic corruptions and fail catastrophically …
Supervised perceptron learning vs unsupervised Hebbian unlearning: Approaching optimal memory retrieval in Hopfield-like networks
The Hebbian unlearning algorithm, ie, an unsupervised local procedure used to improve the
retrieval properties in Hopfield-like neural networks, is numerically compared to a …
retrieval properties in Hopfield-like neural networks, is numerically compared to a …
Model metamers illuminate divergences between biological and artificial neural networks
Deep neural network models of sensory systems are often proposed to learn
representational transformations with invariances like those in the brain. To reveal these …
representational transformations with invariances like those in the brain. To reveal these …
Evolutionary algorithms as an alternative to backpropagation for supervised training of Biophysical Neural Networks and Neural ODEs
Training networks consisting of biophysically accurate neuron models could allow for new
insights into how brain circuits can organize and solve tasks. We begin by analyzing the …
insights into how brain circuits can organize and solve tasks. We begin by analyzing the …
Exploring the perceptual straightness of adversarially robust and biologically-inspired visual representations
Humans have been shown to use a''straightened''encoding to represent the natural visual
world as it evolves in time (H\'enaff et al.~ 2019). In the context of discrete video sequences,'' …
world as it evolves in time (H\'enaff et al.~ 2019). In the context of discrete video sequences,'' …
Complex network effects on the robustness of graph convolutional networks
BA Miller, K Chan, T Eliassi-Rad - Applied Network Science, 2024 - Springer
Vertex classification using graph convolutional networks is susceptible to targeted poisoning
attacks, in which both graph structure and node attributes can be changed in an attempt to …
attacks, in which both graph structure and node attributes can be changed in an attempt to …
Extreme image transformations affect humans and machines differently
G Malik, D Crowder, E Mingolla - Biological Cybernetics, 2023 - Springer
Some recent artificial neural networks (ANNs) claim to model aspects of primate neural and
human performance data. Their success in object recognition is, however, dependent on …
human performance data. Their success in object recognition is, however, dependent on …