Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
The Hintons in your Neural Network: a Quantum Field Theory View of Deep Learning
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 GP interpolation for irregularly sampled inputs instead of a standard MLP input | PNN vs BNN · SyntheticControl, sampling 0.9 | accuracy: 84.66 $\pm$ 0.47 -> 93.33 $\pm$ 1.69 | Helped |
| Use GP interpolation for irregularly sampled inputs instead of a standard MLP input | PNN vs BNN · SyntheticControl, sampling 0.1 | accuracy: 28.33 $\pm$ 8.18 -> 36.00 $\pm$ 5.35 | Helped |
| Use the semiclassical network with symplectic softplus instead of the classical PNN | SPNN vs PNN · SyntheticControl, sampling 0.9 | accuracy: 93.33 $\pm$ 1.69 -> 93.00 $\pm$ 2.82 | No clear effect |
| Use the semiclassical network with symplectic softplus instead of the classical PNN | SPNN vs PNN · SyntheticControl, sampling 0.1 | accuracy: 36.00 $\pm$ 5.35 -> 40.33 $\pm$ 8.05 | No clear effect |
| Subsample input time-series observations | BNN · SyntheticControl, sampling 1 vs 0.1 | accuracy: 94 -> 28.33 $\pm$ 8.18 | Hurt |
Closest papers by meaning
- A Unified Theory of Quantum Neural Network Loss Landscapes (ICLR 2025 · similarity 0.85)
- Quantum Visual Fields with Neural Amplitude Encoding (NeurIPS 2025 · similarity 0.84)
- Quantum Deep Equilibrium Models (NeurIPS 2024 · similarity 0.84)
- Rethink the Role of Deep Learning towards Large-scale Quantum Systems (ICML 2025 · similarity 0.84)
- Exponential Hardness of Optimization from the Locality in Quantum Neural Networks (AAAI 2024 · similarity 0.83)
- Quantum Positional Encodings for Graph Neural Networks (ICML 2024 · similarity 0.83)
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.