In the "Supported Scalarizers" table of the README (also shown on PyPI), two publication links point to unrelated astrophysics papers:
| Scalarizer |
Current link |
Points to |
Correct link |
| COSMOS |
arXiv 2303.04536 |
FUors, EXors and the role of intermediate objects |
arXiv 2103.13392: Scalable Pareto Front Approximation for Deep Multi-Objective Learning, already cited by the COSMOS docstring |
| GeometricMean |
arXiv 1902.08325 |
The Spiral Structure of the Milky Way Galaxy from Radio Observations |
arXiv 1904.08492: MultiNet++: Multi-Stream Feature Aggregation and Geometric Loss Strategy for Multi-Task Learning |
The COSMOS row also uses a title that doesn't match the paper (COSMOS: Enhancing Multi-Objective Optimization with Scalarization); it should use the title of the cited paper.
Relevant lines:
|
| [COSMOS](https://torchjd.org/stable/docs/scalarization/cosmos/) | [COSMOS: Enhancing Multi-Objective Optimization with Scalarization](https://arxiv.org/pdf/2303.04536) | |
|
| [DWA](https://torchjd.org/stable/docs/scalarization/dwa/) | [End-to-End Multi-Task Learning with Attention](https://arxiv.org/pdf/1803.10704) | |
|
| [FAMO](https://torchjd.org/stable/docs/scalarization/famo/) | [FAMO: Fast Adaptive Multitask Optimization](https://arxiv.org/pdf/2306.03792) | |
|
| [GeometricMean](https://torchjd.org/stable/docs/scalarization/geometric_mean/) | [MultiNet++: Multi-Stream Feature Aggregation and Geometric Loss Strategy for Multi-Task Learning](https://arxiv.org/pdf/1902.08325) | |
|
| [IMTL-L](https://torchjd.org/stable/docs/scalarization/imtl_l/) | [Towards Impartial Multi-task Learning](https://discovery.ucl.ac.uk/id/eprint/10120667/) | |
|
| [Mean](https://torchjd.org/stable/docs/scalarization/mean/) | - | |
|
| [PBI](https://torchjd.org/stable/docs/scalarization/pbi/) | [A Decomposition-Based Evolutionary Algorithm for Many Objective Optimization](https://ieeexplore.ieee.org/document/7445185) | |
PBI (to double-check): the README cites IEEE 7445185 (A Decomposition-Based Evolutionary Algorithm for Many Objective Optimization), while the PBI docstring cites MOEA/D (IEEE 4358754), where PBI was introduced. The README should probably cite the same paper as the docstring.
Checklist
In the "Supported Scalarizers" table of the README (also shown on PyPI), two publication links point to unrelated astrophysics papers:
COSMOSdocstringThe COSMOS row also uses a title that doesn't match the paper (COSMOS: Enhancing Multi-Objective Optimization with Scalarization); it should use the title of the cited paper.
Relevant lines:
TorchJD/README.md
Lines 159 to 165 in 9ba45e3
PBI (to double-check): the README cites IEEE 7445185 (A Decomposition-Based Evolutionary Algorithm for Many Objective Optimization), while the
PBIdocstring cites MOEA/D (IEEE 4358754), where PBI was introduced. The README should probably cite the same paper as the docstring.Checklist
PBIdocstring.