Deep Neural Networks (DNNs) have become ubiquitous due to their performance on prediction and classification problems. However, they face a variety of threats as their usage spreads. Model extraction attacks, which steal DNN models, endanger intellectual property, data privacy, and security. Previous research has shown that system-level side channels can be used to leak the architecture of a victim DNN, exacerbating these risks. We propose a novel DNN architecture extraction attack, called EZClone, which us

Deep learning neural nets are making significant advances in artificial intelligence (AI). This is mainly due to their ability to automatically learn and

NeurIPS 2019 Sun Dec 8th through Sat the 14th, 2019 at Vancouver Convention Center Paper ID: 7047 Title: On the Inductive Bias of Neural Tangent Kernels Reviewer 1 Study of the Neural Tangent Kernel (Jacot et al. 2018) is a compelling approach to understanding the dynamics of learning in neural networks. As acknowledged by the authors, while this kernel determines the dynamics only for very wide networks, better understanding of this simplified regime could be a first step towards a fuller understanding of

Experiments and text by Marius Hobbhahn. I would like to thank Jaime Sevilla, Jean-Stanislas Denain, Tamay Besiroglu, Lennart Heim, and Anson Ho for…

Tag: Generative Adversarial Networks (GANs), at Mind Matters

An artificial neural network is a collection of simple interconnected algorithms that process information in response to external input

Skip to content Publications Activities Resources Contact MENU Type: (Select Type) Journal Articles Preprint Book Chapters Conference Articles - Tag: (Select Tag) Algorithms Alzheimer's disease Amygdala amyloid pathology ANNs artificial neural networks Behavior Bioinformatics Biophysical Model CA1 Circuit Model Cluster CMLF Connectivity Decision Making Deep learning Dendrites Dendritic Computations Dendritic Plasticity Dendritic Spikes DG Dopamine EMBO Engram Excitability FKNE FPGA HD-MEAs Hippocampus Inter

For many applications it is critical to know the uncertainty of a neural network's predictions. While a variety of neural network parameter estimation methods have been proposed for uncertainty estimation, they have not been rigorously compared across uncertainty measures. We assess four of these parameter estimation methods to calibrate uncertainty estimation using four different uncertainty measures: entropy, mutual information, aleatoric uncertainty and epistemic uncertainty. We evaluate the calibration

Analyse magnitude-based neural network pruning online. Get per-layer sparsity, compression ratios, and sensitivity curves with R code

Listening in on neural activity is a promising way of restoring the ability to communicate for people whose bodies no longer can. Artificial neural networks are the key middleman in the process