Research map / Optimization / Optimizers
The Devil is in the Detail: A Framework for Macroscopic Prediction via Microscopic Models
Yingxiang Yang, Negar Kiyavash, Le Song, Niao He
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 CSO with GDRS instead of Macro-SL | micro-model with SIR · COVID-19 county data | test loss: 0.086 -> 0.041 | Improved |
| Use CSO with GDRS instead of Macro-SL | micro-model with Neural ODE · COVID-19 county data | test loss: 0.037 -> 0.013 | Improved |
| Use CSO with GDRS instead of Micro-SL | micro-model with Neural ODE · COVID-19 county data | test loss: 0.049 -> 0.013 | Improved |
| Use GDRS instead of biased SGD to solve CSO | SIR micro-model · COVID-19 county data | test loss: 0.044 -> 0.041 | Improved |
| Use GDRS instead of biased SGD to solve CSO | Neural ODE micro-model · COVID-19 county data | test loss: 0.013 -> 0.013 | No clear effect |
| Increase the number of random-search particles m (1 to 1000) (ablation) | GDRS · toy linear regression | l2 test loss convergence speed | Improved |
| Use GDRS instead of biased SGD and Macro-SL | CSO · toy linear regression | l2 test loss | Improved |
| Use Micro-SL instead of GDRS (needs micro labels) | toy linear regression · toy linear regression | l2 test loss | Mixed |
| Use GDRS instead of competing algorithms under noisy training data and small aggregation ratio | CSO · Medicare SynPUF | l2 and KL test loss | Improved |
Closest papers by meaning
- Interpretable Sequence Learning for Covid-19 Forecasting (NeurIPS 2020 · similarity 0.82)
- Scaling Epidemic Inference on Contact Networks: Theory and Algorithms (NeurIPS 2025 · similarity 0.82)
- Learning Macroscopic Dynamics from Partial Microscopic Observations (NeurIPS 2024 · similarity 0.80)
- Reservoir Computing for Short High-Dimensional Time Series: an Application to SARS-CoV-2 Hospitalization Forecast (ICML 2024 · similarity 0.80)
- A Generalizable Physics-Enhanced State Space Model for Long-Term Dynamics Forecasting in Complex Environments (ICML 2025 · similarity 0.79)
- BGDB: Bernoulli-Gaussian Decision Block with Improved Denoising Diffusion Probabilistic Models (AAAI 2025 · similarity 0.79)
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.
