Learning Graph Models for Retrosynthesis Prediction

Authors: Vignesh Ram Somnath, Charlotte Bunne, Connor Coley, Andreas Krause, Regina Barzilay

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

Reproducibility Variable Result LLM Response
Research Type Experimental Our model achieves a top-1 accuracy of 53.7%, outperforming previous template-free and semi-template-based methods.
Researcher Affiliation Collaboration Vignesh Ram Somnath1 Charlotte Bunne1 Connor W. Coley2 Andreas Krause1 Regina Barzilay3 1Department of Computer Science, ETH 2Department of Chemical Engineering, MIT 3Computer Science and Artificial Intelligence Lab, MIT 1{vsomnath, bunnec, krausea}@ethz.ch, 2ccoley@mit.edu, 3regina@csail.mit.edu
Pseudocode No The paper describes the model architecture and processes in text and mathematical equations, but does not contain structured pseudocode or algorithm blocks.
Open Source Code No The paper mentions a GitHub link for a baseline model (RETROXPERT) but does not provide a link or explicit statement about the availability of the source code for their own method, GRAPHRETRO.
Open Datasets Yes We use the benchmark dataset USPTO-50k [Schneider et al., 2016] for all our experiments.
Dataset Splits Yes We use the benchmark dataset USPTO-50k [Schneider et al., 2016] for all our experiments. We use the same dataset version and splits as provided by [Dai et al., 2019].
Hardware Specification No The acknowledgements section states: "We thank the Leonhard scientific computing cluster at ETH Zürich for providing computational resources." This is a general statement and does not include specific hardware details like GPU/CPU models or memory.
Software Dependencies No The paper mentions using a "message passing network (MPN)" and a "Transformer architecture", but it does not specify software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow versions, or specific library versions).
Experiment Setup No The paper describes the model design, training objective, and inference procedure, but it does not explicitly provide specific hyperparameter values (e.g., learning rate, batch size, number of epochs) or other detailed system-level training settings.