An online passive-aggressive algorithm for difference-of-squares classification

Authors: Lawrence Saul

NeurIPS 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental 4 Experimental results We experimented on the INFIMNIST data set of handwritten digit images [7], a purposefully constructed superset of the original MNIST data set [73].
Researcher Affiliation Academia Lawrence K. Saul Department of Computer Science and Engineering University of California, San Diego 9500 Gilman Drive, Mail Code 0404 La Jolla, CA 92093-0404 saul@cs.ucsd.edu
Pseudocode No The paper describes algorithms using mathematical equations and text, but does not provide structured pseudocode or algorithm blocks.
Open Source Code No The paper does not state that source code for the described methodology is publicly available, nor does it provide a link to a code repository.
Open Datasets Yes We experimented on the INFIMNIST data set of handwritten digit images [7], a purposefully constructed superset of the original MNIST data set [73].
Dataset Splits No The paper mentions a test set and training examples but does not explicitly state a validation dataset split or its size/percentage.
Hardware Specification Yes In total we purchased several hundred CPU-hours on an externally managed cluster of Intel Xeon Gold 6132 processors.
Software Dependencies No The paper does not provide specific software dependencies with version numbers for its experiments.
Experiment Setup Yes We initialized the parameters of the linear model with zero values and those of the Do S models with small random values. Specifically, we sampled the elements of U and V from a zero-mean normal distribution whose variance was inversely proportional to the number of elements in these matrices.