Neural Relational Inference for Interacting Systems
Authors: Thomas Kipf, Ethan Fetaya, Kuan-Chieh Wang, Max Welling, Richard Zemel
ICML 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In experiments on simulated physical systems, we show that our NRI model can accurately recover ground-truth interactions in an unsupervised manner. We further demonstrate that we can find an interpretable structure and predict complex dynamics in real motion capture and sports tracking data. |
| Researcher Affiliation | Academia | 1University of Amsterdam, Amsterdam, The Netherlands 2University of Toronto, Toronto, Canada 3Vector Institute, Toronto, Canada 4Canadian Institute for Advanced Research, Toronto, Canada. |
| Pseudocode | No | The paper describes the model architecture and equations but does not provide pseudocode or a clearly labeled algorithm block. |
| Open Source Code | Yes | Our implementation uses Py Torch (Paszke et al., 2017) and is available online2. 2https://github.com/ethanfetaya/nri |
| Open Datasets | Yes | The CMU Motion Capture Database (CMU, 2003) is a large collection of motion capture recordings for various tasks (such as walking, running, and dancing) performed by human subjects. We here focus on recorded walking motion data of a single subject (subject #35). |
| Dataset Splits | Yes | We generate 50k training examples, and 10k validation and test examples for all tasks. We split the different walking trials into non-overlapping training (11 trials), validation (4 trials) and test sets (7 trials). We extracted 12k segments from the 2016 season and used 10k, 1k, 1k for training, validation, and testing respectively. |
| Hardware Specification | No | The paper does not explicitly describe the specific hardware (e.g., GPU models, CPU types) used for running the experiments in the main text. |
| Software Dependencies | No | The paper mentions "Our implementation uses Py Torch (Paszke et al., 2017)", but it does not specify a version number for PyTorch or any other software dependencies. |
| Experiment Setup | Yes | Optimization was performed using the Adam algorithm (Kingma & Ba, 2015). We used M = 10 in our experiments. We train our NRI model with an MLP encoder and RNN decoder on this data using 2 or 4 edge types where one edge type is hard-coded as non-edge, i.e. messages are only passed on the other edge types. The encoder is trained on only the first 17 time steps. |