Research map / Learning theory and algorithms / Prediction and decision losses
Gaussian Gated Linear Networks
David Budden, Adam Marblestone, Eren Sezener, Tor Lattimore, Gregory Wayne, Joel Veness
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 G-GLN instead of DO | UCI regression · Boston Housing | test RMSE: 2.97 -> 2.84 | Improved |
| Use G-GLN instead of DO | UCI regression · Concrete Compression Strength | test RMSE: 5.23 -> 5.84 | Hurt |
| Use G-GLN instead of DO | UCI regression · Energy Efficiency | test RMSE: 1.66 -> 1.31 | Improved |
| Use G-GLN instead of PBP | UCI regression · Kin8nm | test RMSE: 0.10 -> 0.09 | Improved |
| Use G-GLN instead of VI | UCI regression · Naval Propulsion | test RMSE: 0.01 -> 0.00 | Improved |
| Use G-GLN instead of DO | UCI regression · Combined Cycle Power Plant | test RMSE: 4.02 -> 3.90 | Improved |
| Use G-GLN instead of DO | UCI regression · Protein Structure | test RMSE: 4.36 -> 3.77 | Improved |
| Use G-GLN instead of DO | UCI regression · Wine Quality Red | test RMSE: 0.62 -> 0.57 | Improved |
| Use G-GLN instead of PBP | UCI regression · Yacht Hydrodynamics | test RMSE: 1.01 -> 3.76 | Hurt |
| Use G-GLN instead of TabNet-L | SARCOS inverse dynamics · SARCOS inverse dynamics | test MSE: 0.14 -> 0.10 | Improved |
| Use G-GLN instead of LinFullPost | continuous-reward contextual bandits · Financial contextual bandit | cumulative reward: 3122±1 -> 3018±3 | Hurt |
| Use G-GLN instead of BootRMS | continuous-reward contextual bandits · Jester contextual bandit | cumulative reward: 3269±4 -> 3301±4 | Improved |
Closest papers by meaning
- Gated Linear Networks (AAAI 2021 · similarity 0.92)
- Theoretical Characterisation of the Gauss Newton Conditioning in Neural Networks (NeurIPS 2024 · similarity 0.86)
- Exact, Tractable Gauss-Newton Optimization in Deep Reversible Architectures Reveal Poor Generalization (NeurIPS 2024 · similarity 0.85)
- Implicit bias of SGD in $L_2$-regularized linear DNNs: One-way jumps from high to low rank (ICLR 2024 · similarity 0.84)
- Asymptotic Analysis of Two-Layer Neural Networks after One Gradient Step under Gaussian Mixtures Data with Structure (ICLR 2025 · similarity 0.84)
- Benign Overfitting and Grokking in ReLU Networks for XOR Cluster Data (ICLR 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.
