We lifted 13 functions out of this paper's own repositories and ran 11 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| minkaixu/egno | canonical | 11 of 13 |
| Function | Status | Where it lives |
|---|---|---|
| aggregate | Ran | minkaixu/egno/model/basic.py code served (permissive licence) · get_code("c13557c4c4b42aff") |
| collector | Ran | minkaixu/egno/utils.py code served (permissive licence) · get_code("d46f2eac5def9db4") |
| collector_simulation | Ran | minkaixu/egno/utils.py code served (permissive licence) · get_code("09a6af5717cfe3e4") |
| collector_simulation_no | Ran | minkaixu/egno/utils.py code served (permissive licence) · get_code("99eee5d4194d3501") |
| default_init | Ran | minkaixu/egno/model/layer_no.py code served (permissive licence) · get_code("04548c6c9fff2b3f") |
| get_timestep_embedding | Ran | minkaixu/egno/model/layer_no.py code served (permissive licence) · get_code("b663f5d919fd45e0") |
| get_velocity_attr | Ran | minkaixu/egno/main_md17_no.py code served (permissive licence) · get_code("efc11ed98d6d5af4") |
| parse_amc | Ran | minkaixu/egno/motion/amc_parser.py code served (permissive licence) · get_code("00b1bc1dee272f18") |
| read_line | Ran | minkaixu/egno/motion/amc_parser.py code served (permissive licence) · get_code("2cafa259d5ab3508") |
| unsorted_segment_mean | Ran | minkaixu/egno/model/basic.py code served (permissive licence) · get_code("bc66f19be57720e7") |
| variance_scaling | Ran | minkaixu/egno/model/layer_no.py code served (permissive licence) · get_code("f34882fce4b22d20") |
| collate_mda | Not yet run | minkaixu/egno/mdanalysis/dataset.py code served (permissive licence) · get_code("7d820c2f22842d1c") |
| collate_mdd | Not yet run | minkaixu/egno/mdanalysis/dataset.py code served (permissive licence) · get_code("950a677b2fa6cb95") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Modeling the complex three-dimensional (3D) dynamics of relational systems is an important problem in the natural sciences, with applications ranging from molecular simulations to particle mechanics. Machine learning methods have achieved good success by learning graph neural networks to model spatial interactions. However, these approaches do not faithfully capture temporal correlations since they only model next-step predictions. In this work, we propose Equivariant Graph Neural Operator (EGNO), a novel and principled method that directly models dynamics as trajectories instead of just next-step prediction. Different from existing methods, EGNO explicitly learns the temporal evolution of 3D dynamics where we formulate the dynamics as a function over time and learn neural operators to approximate it. To capture the temporal correlations while keeping the intrinsic SE(3)-equivariance, we develop equivariant temporal convolutions parameterized in the Fourier space and build EGNO by stacking the Fourier layers over equivariant networks. EGNO is the first operator learning framework that is capable of modeling solution dynamics functions over time while retaining 3D equivariance. Comprehensive experiments in multiple domains, including particle simulations, human motion capture, and molecular dynamics, demonstrate the significantly superior performance of EGNO against existing methods, thanks to the equivariant temporal modeling. Our code is available at https://github.com/MinkaiXu/egno.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2401.11037")
get_code_for_paper("2401.11037")
have("2401.11037")
Connect an agent — have() is free.