Learning Modular Simulations for Homogeneous Systems
Authors: Jayesh Gupta, Sai Vemprala, Ashish Kapoor
NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate the performance of MP-NODE on five different systems: two of them selected to focus on robotic regimes and three of them as representative of standard graph neural network benchmarks, for all of which we generate datasets of trajectories. ... We compare our proposed approach to the one proposed by Sanchez-Gonzalez et al. [31] that generalized different graph based dynamics learning methods, which we refer to as the L2S baseline. |
| Researcher Affiliation | Industry | Jayesh K. Gupta , Sai Vemprala , Ashish Kapoor Microsoft Autonomous Systems and Robotics Research <jayesh.gupta,savempra,akapoor>@microsoft.com |
| Pseudocode | No | The paper does not contain any structured pseudocode or clearly labeled algorithm blocks. |
| Open Source Code | Yes | Source code for our experiments is available at https://github.com/microsoft/MPNODE.jl. |
| Open Datasets | No | We evaluate the performance of MP-NODE on five different systems: two of them selected to focus on robotic regimes and three of them as representative of standard graph neural network benchmarks, for all of which we generate datasets of trajectories. ... We use Lindner et al. [44] to build our reference simulators to generate our datasets. The paper mentions generating datasets but does not provide specific access information (link, DOI, or formal citation for the dataset) to the generated datasets themselves. |
| Dataset Splits | No | The paper mentions 'training-specific details such as hyperparameters, learning rates, optimizers in the appendix' and discusses 'training sets' and 'validation performance' (Table 1), but it does not specify explicit percentages or absolute counts for training, validation, and test data splits, nor does it refer to predefined splits with citations for these specific experiments. |
| Hardware Specification | No | The provided text does not explicitly describe the specific hardware used (e.g., GPU/CPU models, processor types, or memory details) for running the experiments. It only mentions that details are in the appendix. |
| Software Dependencies | No | The paper states 'We implement our method in Julia [35] and make use of the Sci ML ecosystem [36, 37]', but it does not provide specific version numbers for Julia or any components of the Sci ML ecosystem. |
| Experiment Setup | No | The paper mentions that 'training-specific details such as hyperparameters, learning rates, optimizers' are listed in the appendix. These specific details are not present in the provided main text. |