Research map / Continual learning / Online and replay-based
Forgetting to Learn Logic Programs
Andrew Cropper
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 syntactical forgetting to remove duplicate programs from the allowed predicate symbols | Forgetgol · Robot planning, big dataset, 20,000 tasks | percentage of tasks solved: 100% | Improved |
| Use syntactical forgetting rather than remembering all programs | Forgetgol vs Metabias · Robot planning, big dataset, 20,000 tasks | learning time: twice as quick | Improved |
| Use statistical forgetting instead of syntactical forgetting (ablation) | Forgetgol · Robot planning, big dataset | percentage of tasks solved: <1% | No clear effect |
| Use statistical forgetting rather than remembering everything (ablation) | Forgetgol with statistical forgetting vs Metabias · Robot planning | learning performance | No clear effect |
| Remember all learned programs instead of syntactically forgetting duplicates | Metabias vs Forgetgol with syntactical forgetting · Lego, 20,000 tasks | percentage of tasks solved: around 80% vs 100% | Hurt |
| Use syntactical forgetting rather than remembering all programs | Forgetgol vs Metabias · Lego, 20,000 tasks | learning time: three times quicker | Improved |
| Use statistical forgetting instead of syntactical forgetting (ablation) | Forgetgol · Lego | learning performance | No clear effect |
| Increase the number of tasks while retaining all learned programs | Metabias · Lego, after 12,000 tasks | percentage of tasks solved: around 80% | Hurt |
Closest papers by meaning
- A Unified View on Forgetting and Strong Equivalence Notions in Answer Set Programming (AAAI 2024 · similarity 0.84)
- Learning MDL Logic Programs from Noisy Data (AAAI 2024 · similarity 0.84)
- Learning What to Forget: Improving LLM Unlearning via Learned Token-Level Importance (NeurIPS 2026 · similarity 0.84)
- Light-IF: Endowing LLMs with Generalizable Reasoning via Preview and Self-Checking for Complex Instruction Following (AAAI 2026 · similarity 0.84)
- Long-Tail Class Incremental Learning via Independent Sub-prototype Construction (CVPR 2024 · similarity 0.84)
- SEFE: Superficial and Essential Forgetting Eliminator for Multimodal Continual Instruction Tuning (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.
