Research map / Learning theory and algorithms / Sample complexity
Bias-Variance Trade-Off in Hierarchical Probabilistic Models Using Higher-Order Feature Interactions
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 higher-order interactions instead of hidden layers | HBM versus RBM · synthetic dataset, smaller sample sizes | total KL error | Helped |
| use higher-order interactions instead of hidden layers | HBM versus RBM · synthetic dataset, smaller sample sizes | variance | Helped |
| use hidden layers instead of higher-order interactions | RBM versus HBM · synthetic dataset, larger sample sizes | variance | Helped |
| increase the number of model parameters (ablation) | RBM and HBM · synthetic dataset | variance | Hurt |
| increase the number of model parameters (ablation) | HBM · synthetic dataset | variance | Hurt |
| increase the sample size (ablation) | RBM and HBM · synthetic dataset | variance | Helped |
| vary the number of hidden nodes (ablation) | RBM · synthetic dataset | bias | No clear effect |
| vary the interaction order (ablation) | HBM · synthetic dataset | bias | No clear effect |
| increase the sample size (ablation) | RBM and HBM · synthetic dataset | bias | No clear effect |
| use HBM rather than RBM | HBM versus RBM · synthetic dataset, larger sample sizes | total KL error | No clear effect |
Closest papers by meaning
- Selective Inference for Sparse High-Order Interaction Models (ICML 2017 · similarity 0.85)
- Microcanonical Langevin Ensembles: Advancing the Sampling of Bayesian Neural Networks (ICLR 2025 · similarity 0.84)
- BoltzNCE: Learning likelihoods for Boltzmann Generation with Stochastic Interpolants and Noise Contrastive Estimation (NeurIPS 2025 · similarity 0.83)
- Measuring Stochastic Data Complexity with Boltzmann Influence Functions (ICML 2024 · similarity 0.83)
- Batch and match: black-box variational inference with a score-based divergence (ICML 2024 · similarity 0.83)
- Learning Latent Variable Models via Jarzynski-adjusted Langevin Algorithm (NeurIPS 2025 · 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.