Research map / Architectures and efficiency / RNNs and normalization
Equivariant Neural Networks for General Linear Symmetries on Lie Algebras
Chankyo Kim, Sicheng Zhao, Minghan Zhu, Tzu-Yuan Lin, Maani Ghaffari
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add the semisimple ideal to a center-only model (full reductive structure) (ablation) | ReLN (v, log C) · Drone state estimation | ATE (m): 16.27 -> 13.92 | Improved |
| add the central component to a semisimple-only (Lie Neurons style) model (ablation) | ReLN · Drone state estimation | ATE (m): 16.86 -> 16.49 | Improved |
| use log-covariance instead of raw covariance as input | ReLN · Drone state estimation | ATE (m): 16.49 -> 13.92 | Improved |
| add covariance C to velocity input | ReLN · Drone state estimation | ATE (m): 16.85 -> 16.49 | Improved |
| add covariance C to velocity input via eigendecomposition | VN · Drone state estimation | ATE (m): 17.36 -> 191.78 | Hurt |
| replace non-equivariant ResNet with ReLN | ResNet · Drone state estimation | ATE (m): 205.11 -> 16.49 | Improved |
| replace MLP with ReLN under camera rotation | MLP · Platonic solid classification (rotated camera) | accuracy (%): 36.54% -> 99.78% | Improved |
| add SO(3) augmentation to MLP | MLP · Platonic solid classification (in-distribution) | accuracy (%): 95.76% -> 81.47% | Hurt |
| double MLP width | MLP · Platonic solid classification (rotated camera) | accuracy (%): 36.54% -> 36.55% | No clear effect |
| replace Lie Neurons with ReLN | Lie Neurons · sp(4) invariant regression | test MSE: 5.83x10^-4 vs 5.14x10^-4 | No clear effect |
| add Sp(4) conjugation augmentation to MLP 512 | MLP 512 · sp(4) invariant regression, Sp(4)-conjugated test | test MSE: 0.906 -> 0.446 | Improved |
| replace baseline Gaussian-MAE with ReLN encoder/decoder | Gaussian-MAE · ModelNet10 classification, randomly rotated test | accuracy (%): 18.28 -> 95.15 | Improved |
Closest papers by meaning
- Lie Algebra Canonicalization: Equivariant Neural Operators under arbitrary Lie Groups (ICLR 2025 · similarity 0.92)
- Lie Neurons: Adjoint-Equivariant Neural Networks for Semisimple Lie Algebras (ICML 2024 · similarity 0.91)
- Lie Group Decompositions for Equivariant Neural Networks (ICLR 2024 · similarity 0.90)
- Explicit Discovery of Nonlinear Symmetries from Dynamic Data (ICML 2025 · similarity 0.88)
- A Tale of Two Symmetries: Exploring the Loss Landscape of Equivariant Models (NeurIPS 2025 · similarity 0.87)
- A Characterization Theorem for Equivariant Networks with Point-wise Activations (ICLR 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
