Software & Data Downloads — G-RepsNets
Group Representation Networks for general and efficient construction of equivariant neural networks using equivariant tensor polynomials and tensor mixing.
This is the code for the TMLR 2025 publication G-RepsNet: A Fast and General Construction of Equivariant Networks for Arbitrary Matrix Groups. Given input representations and the group of interest, the work presents a general and efficient construction of equivariant neural networks using equivariant tensor polynomials and tensor mixing.
-
MERL Contacts
-
Related Publications
- , "G-RepsNet: A Lightweight Construction of Equivariant Net- works for Arbitrary Matrix Groups", Transactions on Machine Learning Research (TMLR), May 2025.
BibTeX TR2025-056 PDF Software- @article{Basu2025may,
- author = {Basu, Sourya and Lohit, Suhas and Brand, Matthew},
- title = {{G-RepsNet: A Lightweight Construction of Equivariant Net- works for Arbitrary Matrix Groups}},
- journal = {Transactions on Machine Learning Research (TMLR)},
- year = 2025,
- month = may,
- issn = {2835-8856},
- url = {https://www.merl.com/publications/TR2025-056}
- }
- , "G-RepsNet: A Lightweight Construction of Equivariant Net- works for Arbitrary Matrix Groups", Transactions on Machine Learning Research (TMLR), May 2025.
Software & Data Downloads
Access software at https://github.com/merlresearch/G-RepsNets.