Research map / Architectures and efficiency / CNN design
Deep Signature Transforms
Patrick Kidger, Patric Bonnier, Imanol Perez Arribas, Cristopher Salvi, Terry Lyons
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 |
|---|---|---|---|
| Use a learned transformation before the signature instead of a fixed signature feature transform | DeepSigNet vs Neural-Sig · Fractional Brownian motion Hurst estimation test set | MSE: 1.1 \times 10^{-2} -> 2.1 \times 10^{-4} | Improved |
| Use three smaller Neural-Lift-Signature blocks instead of one large block | DeeperSigNet vs DeepSigNet · Fractional Brownian motion Hurst estimation test set | MSE: 2.1 \times 10^{-4} -> 1.6 \times 10^{-4} | Improved |
| Use deep signature memory instead of a comparable recurrent network | Deep signature RL model vs RNN · Partially observed Mountain Car | episodes to consistently solve | Improved |
| Train the generator using the signature-kernel MMD objective | Deep signature generative model · Ornstein–Uhlenbeck process generation | MMD loss: 6.6 \times 10^{-4} | Improved |
Closest papers by meaning
- Learning with Expected Signatures: Theory and Applications (ICML 2025 · similarity 0.84)
- On the Optimization and Generalization of Two-layer Transformers with Sign Gradient Descent (ICLR 2025 · similarity 0.82)
- Beyond Slow Signs in High-fidelity Model Extraction (NeurIPS 2024 · similarity 0.82)
- Rough Transformers: Lightweight and Continuous Time Series Modelling through Signature Patching (NeurIPS 2024 · similarity 0.82)
- Real-Fake: Effective Training Data Synthesis Through Distribution Matching (ICLR 2024 · similarity 0.82)
- Probabilistic Signature Inversion: Learning Conditional Distributions from Truncated Signatures (NeurIPS 2026 · similarity 0.81)
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.
