Tails of Lipschitz Triangular Flows

Authors: Priyank Jaini, Ivan Kobyzev, Yaoliang Yu, Marcus Brubaker

ICML 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We perform several synthetic and real-world experiments to compliment our theoretical findings.
Researcher Affiliation Collaboration 1Univesity of Waterloo, Waterloo, Canada 2Vector Institute, Toronto, Canada 3Borealis AI 4York University, Toronto, Canada.
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide concrete access to source code for the methodology described. There is no specific repository link, explicit code release statement, or mention of code in supplementary materials.
Open Datasets Yes Lastly, we replicate density estimation experiments on benchmark datasets popularly used to measure performance of flows and autoregressive models.
Dataset Splits Yes We divided the dataset into training-validation-testing in the ratio 2:1:1.
Hardware Specification Yes We thank NVIDIA Corporation (the data science grant) for donating two Titan V GPUs that enabled in part the computation in this work.
Software Dependencies No The paper mentions using "Adam (Kingma & Ba, 2014)" but does not specify version numbers for any software dependencies (e.g., Python, PyTorch, TensorFlow, or Adam itself).
Experiment Setup Yes We trained the model using Adam (Kingma & Ba, 2014) for 40 epochs with a batch size of 128 and learning rate of 10-3.