Research map / Architectures and efficiency / RNNs and normalization
Modelling Autobiographical Memory Loss across Life Span
Di Wang, Ah-Hwee Tan, Chunyan Miao, Ahmed A. Moustafa
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 |
|---|---|---|---|
| polynomial instead of linear extrapolation of parameters (ablation) | AM-ART · Predicting 70s recall | RMSE: 0.0732 -> 0.0516 | Improved |
| polynomial instead of linear extrapolation (ablation) | AM-ART · Predicting 60s recall | RMSE: 0.0495 -> 0.0363 | Improved |
| polynomial instead of linear extrapolation (ablation) | AM-ART · Predicting 50s recall | RMSE: 0.0312 -> 0.0294 | Improved |
| polynomial instead of linear extrapolation (ablation) | AM-ART · Predicting 30s recall | RMSE: 0.0119 -> 0.0177 | Hurt |
| polynomial instead of linear extrapolation (ablation) | AM-ART · Predicting 40s recall | RMSE: 0.0141 -> 0.0169 | Hurt |
| learnt-parameter extrapolation instead of random parameter values in range (ablation) | AM-ART (linear) · Predicting 30s-70s recall | RMSE: 0.0231 -> 0.0119 (30s); 0.0822 -> 0.0732 (70s) | Improved |
| polynomial extrapolation instead of random parameter values (ablation) | AM-ART · Predicting 70s recall | RMSE: 0.0822 -> 0.0516 | Improved |
| polynomial extrapolation instead of random parameter values (ablation) | AM-ART · Predicting 30s recall | RMSE: 0.0231 -> 0.0177 | Improved |
| polynomial degree 2 instead of degree 1 for extrapolating zeta++ in 0s (ablation) | AM-ART · Emotional coefficient estimation | zeta++ validity | Improved |
| emulated recall vs human survey curves | AM-ART · Berntsen and Rubin life-span recall | correlation: 0.793±0.166 | Improved |
| emulation compared to an independent population (Rubin and Schulkind) | AM-ART · 20s recall ratio | recall ratio: 0.830 -> 0.857 | Improved |
| emulation compared to independent population | AM-ART · 70s recall ratio | recall ratio: 0.476 -> 0.386 | Mixed |
Closest papers by meaning
- Do Your Best and Get Enough Rest for Continual Learning (CVPR 2025 · similarity 0.84)
- Predictive Attractor Models (NeurIPS 2024 · similarity 0.84)
- Factual recall in linear associative memories: sharp asymptotics and mechanistic insights (NeurIPS 2026 · similarity 0.84)
- Memoria: Resolving Fateful Forgetting Problem through Human-Inspired Memory Architecture (ICML 2024 · similarity 0.84)
- BrainLM: A foundation model for brain activity recordings (ICLR 2024 · similarity 0.84)
- Human-inspired Episodic Memory for Infinite Context LLMs (ICLR 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.
