Cooperative Distribution Alignment via JSD Upper Bound

Authors: Wonwoong Cho, ZIYU GONG, David I. Inouye

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We show empirical results on both simulated and real-world datasets to demonstrate the benefits of our approach.
Researcher Affiliation Academia Wonwoong Cho Purdue University cho436@purdue.edu Purdue University gong123@purdue.edu David I. Inouye Purdue University dinouye@purdue.edu Elmore Family School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN
Pseudocode Yes Algorithm 1 Training algorithm for AUB
Open Source Code Yes Code is available at https://github.com/inouye-lab/alignment-upper-bound.
Open Datasets Yes In both experiments, we used four UCI tabular datasets [29] (MINIBOONE, GAS, HEPMASS, and POWER), following the same preprocessing as the MAF paper [30]. ... We perform an image translation task on MNIST dataset4 [31] ... Domain Adaptation on USPS-MNIST dataset
Dataset Splits Yes Train, validation, and test sets are 80%, 10%, and 10% of the data respectively.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running experiments.
Software Dependencies No The paper mentions software components like 'Real NVP' but does not provide specific version numbers for any software dependencies.
Experiment Setup No The paper mentions 'Implementation details are provided in ??' which suggests they are outside the main text. Algorithm 1 lists some general parameters like 'batch size; learning rate η; maximum epoch Emax' but no specific values are given in the main body of the paper.