Research map / Probabilistic and scientific modeling / Molecules and proteins
WIET: Harmonizing Group-aware Model Weighting and Worker Allocation for Ensemble Temporal Prediction MaaS
Binbin Feng, Shikun He, Yingxin Wang, Pengwei Wang, Xiang Gao, Zhijun Ding
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 |
|---|---|---|---|
| add group temporal locality weighting to the random forest weighting model (ablation) | MNW · Temperature | MSE: MNW 1.064 times WIET MSE (RF) | Improved |
| add group temporal locality weighting | MNW · Stock | MAE: 15.6% and 25.2% MAE reduction vs MNW | Improved |
| add group temporal locality weighting (SVM classifier) (ablation) | MNW · Temperature | MSE: 1.536 ratio MNW/WIET | Improved |
| add group temporal locality weighting (XGBoost classifier) (ablation) | MNW · Stock | MSE: 1.236 ratio MNW/WIET | Improved |
| add group temporal locality weighting (ablation) | MNW · Temperature | R2: 0.923 to 0.946 ratio MNW/WIET across classifiers | Mixed |
| add quantum-inspired layer to simulated annealing worker mapping (ablation) | WIET (w/o Quantum) · Temperature, Q=15 | latency (ms): 124.73 -> 122.61 | Improved |
| add quantum-inspired layer to simulated annealing worker mapping (ablation) | WIET (w/o Quantum) · Temperature, Q=25 | latency (ms): 468.97 -> 460.23 | Improved |
| add quantum-inspired layer to simulated annealing worker mapping (ablation) | WIET (w/o Quantum) · Stock, Q=25 | latency (ms): 43.28 -> 41.96 | Improved |
| add quantum-inspired layer to simulated annealing worker mapping (ablation) | WIET (w/o Quantum) · Stock, Q=35 | latency (ms): 39.39 -> 38.22 | Improved |
| use WIET worker allocation instead of Kubernetes, Knative, HEAW, SLAM, ChunkFunc | WIET · Temperature and Stock | CPU, memory, response time: 0.94-3.29 cores, 0.38-1.90 GB, 2.47-195.62 ms | Improved |
Closest papers by meaning
- Algorithms with Calibrated Machine Learning Predictions (ICML 2025 · similarity 0.81)
- Swimba: Switch Mamba Model Scales State Space Models (NeurIPS 2026 · similarity 0.81)
- Breaking Silos: Adaptive Model Fusion Unlocks Better Time Series Forecasting (ICML 2025 · similarity 0.81)
- FastPERT: Towards Fast Microservice Application Latency Prediction via Structural Inductive Bias over PERT Networks (AAAI 2025 · similarity 0.81)
- Optimal Aggregation of LLM and PRM Signals for Efficient Test-Time Scaling (ICLR 2026 · similarity 0.81)
- IW-GAE: Importance weighted group accuracy estimation for improved calibration and model selection in unsupervised domain adaptation (ICML 2024 · similarity 0.80)
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.
