Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
Large Margin Deep Networks for Classification
Authors: Gamaleldin Elsayed, Dilip Krishnan, Hossein Mobahi, Kevin Regan, Samy Bengio
NeurIPS 2018 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We demonstrate that the decision boundary obtained by our loss has nice properties compared to standard classification loss functions. Specifically, we show improved empirical results on the MNIST, CIFAR-10 and Image Net datasets on multiple tasks: generalization from small training sets, corrupted labels, and robustness against adversarial perturbations. |
| Researcher Affiliation | Industry | Gamaleldin F. Elsayed Google Research Dilip Krishnan Google Research Hossein Mobahi Google Research Kevin Regan Google Research Samy Bengio Google Research |
| Pseudocode | No | The paper presents mathematical formulations but does not include any structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | Code for the large margin loss function is released at https://github.com/google-research/ google-research/tree/master/large_margin |
| Open Datasets | Yes | Specifically, we show improved empirical results on the MNIST, CIFAR-10 and Image Net datasets on multiple tasks |
| Dataset Splits | Yes | We hold out 5, 000 samples of the training set as a validation set, and the remaining 55, 000 samples are used for training. |
| Hardware Specification | Yes | measured on the same NVIDIA p100 GPU |
| Software Dependencies | No | Our code was written in Tensorflow (Abadi et al., 2016). (No version number provided for TensorFlow or other software components.) |
| Experiment Setup | Yes | We train a 4 hidden-layer model with 2 convolutional layers and 2 fully connected layers, with rectified linear unit (Re Lu) activation functions, and a softmax output layer. The first baseline model uses a cross-entropy loss function, trained with stochastic gradient descent optimization with momentum and learning rate decay. |