Research map / Architectures and efficiency / RNNs and normalization
Expressive power of recurrent neural networks
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| use a TT-Network instead of a CP-Network | classification on CIFAR-10 without regularizers · CIFAR-10 | test accuracy: 0.2 -> 0.45 | Helped |
| apply dropout 0.8 to each patch (ablation) | TT-Network and CP-Network · MNIST | test accuracy: 0.95 -> 0.98 | Helped |
| use equal intermediate TT-cores (ablation) | TT-Network · Equal-core TT tensors, d=6, n ranging from 2 to 10 and r ranging from 2 to 20 | CP-rank lower-bound verification | Helped |
| increase CP-Network rank (ablation) | CP-Network · CIFAR-10 | training accuracy | Mixed |
Closest papers by meaning
- Boosting Dilated Convolutional Networks with Mixed Tensor Decompositions (ICLR 2018 · similarity 0.89)
- The Expressive Power of Neural Networks: A View from the Width (NeurIPS 2017 · similarity 0.89)
- Separation Power of Equivariant Neural Networks (ICLR 2025 · similarity 0.86)
- Convolutions and More as Einsum: A Tensor Network Perspective with Advances for Second-Order Methods (NeurIPS 2024 · similarity 0.85)
- When Do Transformers Outperform Feedforward and Recurrent Networks? A Statistical Perspective (NeurIPS 2025 · similarity 0.85)
- On the Expressiveness of Rational ReLU Neural Networks With Bounded Depth (ICLR 2025 · similarity 0.85)
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.