Learning to Convolve: A Generalized Weight-Tying Approach

Authors: Nichita Diaconu, Daniel Worrall

ICML 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We present our results on some basic benchmarks. We demonstrate competitive classification performance on the CIFAR-10 image recognition dataset, and compare the equivariance properties of activations from different models.
Researcher Affiliation Collaboration Nichita Diaconu 1 * Daniel Worrall 1 * ... 1Philips Lab, University of Amsterdam, Netherlands.
Pseudocode Yes Algorithm 1 Task-specific training using pre-trained basis; Algorithm 2 The basis is pretrained offline
Open Source Code No The paper does not provide any specific links or explicit statements about releasing source code for the described methodology.
Open Datasets Yes We present our results on some basic benchmarks... experiments on MNIST and CIFAR-10. ...Krizhevsky, A. Learning multiple layers of features from tiny images. Technical report, 2009.
Dataset Splits No The paper mentions 'Validation Set Angle' and 'Validation Set Error (%)' in figures, implying a validation set was used, but it does not specify exact split percentages, absolute sample counts, or explicit references to predefined splits for reproducibility.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running its experiments.
Software Dependencies No The paper mentions using 'AMSGrad variant of Adam' but does not specify software dependencies with version numbers (e.g., Python 3.x, PyTorch x.x).
Experiment Setup Yes We used the AMSGrad variant of Adam (Reddi et al., 2018) as the optimizer and we trained for 100 epochs at learning rate 10 3 and weight decay 10 6. We use a minibatch size of 100. For data augmentation, we use random flips, color normalization and random translations of at most 4 pixels.