On Slicing Optimality for Mutual Information

Authors: Ammar Fayad, Majd Ibrahim

NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Through comprehensive experiments in benchmark domains, we demonstrate significant gains in our information measure than state-of-the-art baselines.
Researcher Affiliation Academia Ammar Fayad MIT Majd Ibrahim HIAST afayad@mit.edu
Pseudocode Yes We refer to Appendix C for the pseudocode of SI W.
Open Source Code No We refer to Appendix B for further analysis experiments and Appendix D for implementation details and hyperparameters provided at https://bit.ly/3fo Lke2. The paper does not explicitly state that source code is provided, only "implementation details and hyperparameters."
Open Datasets Yes We test these three methods along with Bi GAN (Donahue et al., 2016) on the STL-10 (Coates et al., 2011) and CIFAR10 (Krizhevsky et al., 2009) datasets, which consist of high-dimensional images.
Dataset Splits No The paper uses datasets like STL-10 and CIFAR10 and performs experiments but does not explicitly provide details on how the datasets were split into training, validation, and test sets or mention specific percentages/counts for reproduction.
Hardware Specification No The paper does not explicitly describe the specific hardware used for its experiments, such as GPU or CPU models, or cloud computing instance types.
Software Dependencies No The paper does not explicitly provide specific software dependencies, such as library names with version numbers, required to replicate the experiments.
Experiment Setup Yes We refer to Appendix B for further analysis experiments and Appendix D for implementation details and hyperparameters provided at https://bit.ly/3fo Lke2.