With Friends Like These, Who Needs Adversaries?

Authors: Saumya Jetley, Nicholas Lord, Philip Torr

NeurIPS 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Via a novel experimental analysis, we illustrate some facts about deep convolutional networks for image classification that shed new light on their behaviour and how it connects to the problem of adversaries.
Researcher Affiliation Collaboration 1Department of Engineering Science, University of Oxford 2Oxford Research Group, Five AI Ltd.
Pseudocode Yes Algorithm 1 Computes mean principal directions and principal curvatures for a net s image-space decision surface.
Open Source Code Yes Source code for replicating all experiments is provided at https://github.com/torrvision/whoneedsadversaries.
Open Datasets Yes CIFAR10-NIN, MNIST-Le Net, CIFAR10-Alex Net, CIFAR100-VGG, IMAGENET-Alex Net, Image Net [26]
Dataset Splits No The paper mentions 'training sets' and 'test data' but does not specify a distinct validation set split or explicit percentages for any splits.
Hardware Specification No The paper does not provide specific details about the hardware (e.g., GPU/CPU models, memory) used for running the experiments.
Software Dependencies No The paper does not list specific software dependencies with version numbers (e.g., library or solver names with versions) required to replicate the experiment.
Experiment Setup No The paper describes experimental procedures and analyses (e.g., perturbation generation, SVD on gradients) but does not explicitly provide specific experimental setup details such as hyperparameters (e.g., learning rates, batch sizes, epochs) for the models or perturbation generation processes beyond general algorithmic descriptions.