XOR-CD: Linearly Convergent Constrained Structure Generation

Authors: Fan Ding, Jianzhu Ma, Jinbo Xu, Yexiang Xue

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

Reproducibility Variable Result LLM Response
Research Type Experimental Our real-world experiments on datadriven experimental design, dispatching route generation, and sequence-based protein homology detection demonstrate the superior performance of XOR-CD compared to baseline approaches in generating valid structures as well as capturing the inductive bias in the training set.
Researcher Affiliation Collaboration 1Department of Computer Science, Purdue University, West Lafayette, USA 2Institute for Artificial Intelligence, Peking University, Beijing, China 3Toyota Technological Institute at Chicago, Illinois, USA.
Pseudocode Yes The detailed procedure of XOR-CD is shown in Algorithm 1. ... Algorithm 1 XOR-CD
Open Source Code No The paper does not provide an explicit statement or a link to open-source code for the described methodology.
Open Datasets Yes We constructed the training set from the PDB40 dataset (Wu & Xu, 2020).
Dataset Splits Yes The test set is made up with 50 randomly sampled sequences from the PDB40 dataset separated from the training set.
Hardware Specification No The paper does not provide any specific hardware details such as CPU/GPU models, memory, or cloud instance types used for the experiments.
Software Dependencies No The paper mentions general software like Python but does not provide specific version numbers for libraries or tools used in their experimental setup.
Experiment Setup Yes Learning rate is fixed as 0.1 and total number of epochs T is 500. There is also a timeout of 10 hours for all algorithms. We also set both M and K to be 100, and parameters for XOR-Sampling were kept the same as in (Ermon et al., 2013b).