Research map / Label-efficient and robust learning / Recognition and long-tail
Backward-Compatible Prediction Updates: A Probabilistic Approach
Frederik Träuble, Julius Von Kügelgen, Matthäus Kleindessner, Francesco Locatello, Bernhard Schölkopf, Peter Vincent Gehler
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 CR 10 update rule instead of Replace | posterior-based updates · ImageNet->ImageNet, budget 100% | cumulative negative flips: 24214 -> 2112 | Improved |
| use CR 10 instead of Replace at budget 100% | posterior-based updates · ImageNet->ImageNet, budget 100% | final accuracy (%): 79.2 -> 77 | Hurt |
| use CR 10 instead of Replace at budget 10% with entropy selection | posterior-based updates · ImageNet->ImageNet, budget 10% | cumulative negative flips: 2468 -> 515 | Improved |
| use CR 10 instead of Replace at budget 10% with entropy selection | posterior-based updates · ImageNet->ImageNet, budget 10% | final accuracy (%): 76.1 -> 75.2 | Hurt |
| select re-evaluation samples by posterior entropy instead of randomly (ablation) | Replace strategy · ImageNet->ImageNet, budget 10% | final accuracy (%): 71.3 -> 76.1 | Improved |
| select re-evaluation samples by posterior entropy instead of randomly (ablation) | Replace strategy · ImageNet->ImageNet, budget 10% | cumulative negative flips: 1958.4 -> 2468 | Hurt |
| use CR 10 instead of Replace on ObjectNet at budget 10% | posterior-based updates · ImageNet->ObjectNet, budget 10% | final accuracy (%): 23.7 -> 17.2 | Hurt |
| use CR 10 instead of Replace on ObjectNet at budget 100% | posterior-based updates · ImageNet->ObjectNet, budget 100% | cumulative negative flips: 16669 -> 1084 | Improved |
| estimate full Laplace-smoothed confusion matrices instead of diagonal only (ablation) | posterior-based updates · ImageNet and ObjectNet | negative flips | Improved |
| estimate full confusion matrices instead of diagonal only (ablation) | conservative CR strategies at small budget · ImageNet and ObjectNet | accuracy gain | Mixed |
Closest papers by meaning
- When to retrain a machine learning model (ICML 2025 · similarity 0.86)
- Reproduce, Replicate, Reevaluate. The Long but Safe Way to Extend Machine Learning Methods (AAAI 2024 · similarity 0.86)
- RCCDA: Adaptive Model Updates in the Presence of Concept Drift under a Constrained Resource Budget (NeurIPS 2025 · similarity 0.85)
- Algorithms with Calibrated Machine Learning Predictions (ICML 2025 · similarity 0.83)
- Towards Establishing Guaranteed Error for Learned Database Operations (ICLR 2024 · similarity 0.83)
- Quilt: Robust Data Segment Selection against Concept Drifts (AAAI 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.
