Accurately Solving Rod Dynamics with Graph Learning

Authors: Han Shao, Tassilo Kugelstadt, Torsten Hädrich, Wojtek Palubicki, Jan Bender, Soeren Pirk, Dominik L Michels

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

Reproducibility Variable Result LLM Response
Research Type Experimental We generate training and validation datasets based on two scenarios: an initially straight bending rod and an elastic helix each fixed at one end oscillating under the influence of gravity. The specification of these datasets is provided in Table 1.
Researcher Affiliation Collaboration KAUST han.shao@kaust.edu.sa Tassilo Kugelstadt RWTH Aachen University kugelstadt@cs.rwth-aachen.de Torsten H adrich KAUST torsten.hadrich@kaust.edu.sa Wojciech Pałubicki AMU wp06@amu.edu.pl Jan Bender RWTH Aachen University bender@cs.rwth-aachen.de S oren Pirk Google Research pirk@google.com Dominik L. Michels KAUST dominik.michels@kaust.edu.sa
Pseudocode Yes Algorithm 1 Numerical integration procedure updating pi,t 7! pi,t+ t and υi,t 7! υi,t+ t.
Open Source Code Yes Did you include the code, data, and instructions needed to reproduce the main experimental results (either in the supplemental material or as a URL)? [Yes] This can be found in the supplement material.
Open Datasets No The paper states: 'We generate training and validation datasets based on two scenarios: an initially straight bending rod and an elastic helix each fixed at one end oscillating under the influence of gravity.' While it mentions data is in supplementary material, it doesn't provide a specific link, DOI, or formal citation for this generated dataset to be considered publicly available in the sense of the question.
Dataset Splits Yes Table 1: Specification of training and validation datasets for the two scenarios of an initially straight bending rod (top) and an elastic helix (bottom). The datasets are comprised of a number of data points (left) each describing the rod s dynamics within t 2 [0 s, 50 t] discretized with a time step size of t = 0.02 s.
Hardware Specification Yes The training is performed on an NVIDIA Tesla V100 GPU.
Software Dependencies No The paper states: 'The PBD code is written in C++, while the COPINGNet is implemented in Py Torch.' No specific version numbers for PyTorch or other libraries are provided.
Experiment Setup Yes A constant learning rate of = 0.001 was used and a mean square loss function was employed.