Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
BATIS: Bayesian Approaches for Targeted Improvement of Species Distribution Models
Catherine Villeneuve, Benjamin Akera, Mélisande Teng, David Rolnick
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 |
|---|---|---|---|
| update the prior with five checklists using MVN | ResNet-18 · Kenya | MAE [1e-2]: 2.04 -> 1.52 | Improved |
| update the prior with five checklists using MVN | ResNet-18 · South Africa | Top-10: 37.01 -> 60.19 | Improved |
| use aleatoric (MVN/HetReg) rather than epistemic-only (Deep Ensembles) uncertainty (ablation) | ResNet-18 · South Africa | Top-10 after 5 updates: 48.06 (DE) -> 60.19 (MVN) | Improved |
| add epistemic estimation (HetReg) on top of aleatoric MVN (ablation) | ResNet-18 · South Africa | Top-10 after 5 updates: 60.19 (MVN) vs 60.10 (HetReg) | No clear effect |
| update with MC Dropout prior | ResNet-18 · South Africa | MAE [1e-2]: 2.47 -> 2.47 | No clear effect |
| update with Shallow Ensemble prior | ResNet-18 · South Africa | Top-10: 41.90 -> 42.80 | Improved |
| use fixed maximum variance prior (FV) instead of learned variance | ResNet-18 · South Africa | MAE [1e-2]: 2.47 -> 1.65 | Improved |
| use historical variance from 5 past checklists | ResNet-18 · South Africa | MAE [1e-2]: 2.47 -> 2.06 | Improved |
| update with MVN prior | ResNet-18 · Kenya | Top-k: 66.38 -> 61.46 | Hurt |
| first Bayesian update with large prior variance | uncertainty-aware SDMs · South Africa | MAE/MSE/Top-10: initial drop before recovery | Hurt |
Closest papers by meaning
- DivShift: Exploring Domain-Specific Distribution Shifts in Large-Scale, Volunteer-Collected Biodiversity Datasets (AAAI 2025 · similarity 0.86)
- GeoPlant: Spatial Plant Species Prediction Dataset (NeurIPS 2024 · similarity 0.85)
- Combining Observational Data and Language for Species Range Estimation (NeurIPS 2024 · similarity 0.85)
- Feedforward Few-shot Species Range Estimation (ICML 2025 · similarity 0.85)
- BioTrove: A Large Curated Image Dataset Enabling AI for Biodiversity (NeurIPS 2024 · similarity 0.83)
- Spatial Clustering of Citizen Science Data Improves Downstream Species Distribution Models (AAAI 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.
