Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Approximate Bayesian Computation via Classification
Yuexi Wang, Tetsuya Kaji, Veronika Rockova
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 exponential kernel aggregation instead of accept/reject (ablation) | RF-ABC · M/G/1 queuing theta2 | 95% CI width: 4.304 -> 1.473 | Improved |
| use exponential kernel aggregation instead of accept/reject (ablation) | RF-ABC · M/G/1 queuing theta1 | 95% CI width: 2.044 -> 0.959 | Improved |
| use exponential kernel aggregation instead of accept/reject (ablation) | RF-ABC · M/G/1 queuing theta1 | squared error: 0.008 -> 0.035 | Hurt |
| use exponential kernel aggregation instead of accept/reject (ablation) | RF-ABC · Lotka-Volterra theta1 | 95% CI width: 1.116 -> 0.354 | Improved |
| use exponential kernel aggregation instead of accept/reject (ablation) | LRD-ABC · Lotka-Volterra theta1 | squared error: 42.949 -> 0.066 | Improved |
| use exponential kernel aggregation instead of accept/reject (ablation) | kNN-ABC · M/G/1 queuing theta1 | squared error: 0.525 -> 1.057 | Hurt |
| use random forest discriminator instead of penalized logistic discriminator | KL-ABC · M/G/1 queuing theta1 | squared error: 0.197 -> 0.008 | Improved |
| use random forest discriminator instead of penalized logistic discriminator | KL-ABC · Lotka-Volterra theta1 | squared error: 42.949 -> 0.118 | Improved |
| use random forest discriminator instead of penalized logistic discriminator | KL-ABC · stock volatility | accuracy of estimation | Hurt |
| use classifier-based KL estimate instead of raw classification accuracy | ABC discrepancy · M/G/1 queuing theta1 | squared error: 0.191 -> 0.008 | Improved |
| use classifier-based KL estimate instead of 2-Wasserstein distance | ABC discrepancy · M/G/1 queuing theta1 | squared error: 0.595 -> 0.008 | Improved |
Closest papers by meaning
- DR-ABC: Approximate Bayesian Computation with Kernel-Based Distribution Regression (ICML 2016 · similarity 0.88)
- Approximate Bayesian Computation with Domain Expert in the Loop (ICML 2022 · similarity 0.88)
- Scalable Bayesian Learning with posteriors (ICLR 2025 · similarity 0.85)
- Do Bayesian Neural Networks Actually Behave Like Bayesian Models? (ICML 2025 · similarity 0.84)
- Learning Likelihood-Free Reference Priors (ICML 2025 · similarity 0.83)
- Decision Making under the Exponential Family: Distributionally Robust Optimisation with Bayesian Ambiguity Sets (ICML 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.
