Research map / Architectures and efficiency / RNNs and normalization
Data Sharing and Compression for Cooperative Networked Control
Jiangnan Cheng, Marco Pavone, Sachin Katti, Sandeep P. Chinchali, Ao Tang
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 |
|---|---|---|---|
| Co-design compressed forecasts with the controller's task objective instead of optimizing only prediction MSE | MPC on IoT sensor data · IoT sensor data | control cost: within 5 % of the optimal cost; 15× | Improved |
| Co-design compressed forecasts with the controller's task objective instead of optimizing only prediction MSE | MPC on cell traffic data · Cell traffic data | control cost: within 5 % of the optimal cost; 15× | Improved |
| Co-design compressed forecasts with the controller's task objective instead of optimizing only prediction MSE | MPC on battery data · Battery data | control cost: within 5 % of the optimal cost; 96× | Improved |
| Use task-agnostic MSE forecasting instead of task-aware co-design | MPC on IoT, cell traffic, and battery data · IoT, cell traffic, and battery data | control cost at the same bottleneck size: at least 25% more control cost | Hurt |
| Use task-agnostic MSE forecasting rather than task-aware co-design | MPC on IoT sensor data · IoT sensor data | data transmitted to reach within 5% of optimal cost: 88 % less data with our approach | Hurt |
| Use task-agnostic MSE forecasting rather than task-aware co-design | MPC on battery data · Battery data | data transmitted to reach within 5% of optimal cost: 82 % less data with our approach | Hurt |
| Use task-agnostic MSE forecasting rather than task-aware co-design | MPC on cell traffic data · Cell traffic data | control cost relative to optimal: 100% more cost than the optimal | Hurt |
| Add prediction-error regularization to fully task-aware co-design (weighted approach) (ablation) | Co-designed MPC forecaster · IoT, cell traffic, and battery data | forecasting error in the near future | Improved |
| Add prediction-error regularization to fully task-aware co-design (weighted approach) (ablation) | Co-designed MPC forecaster · IoT, cell traffic, and battery data | bottleneck size needed for control performance: marginally larger representation | Hurt |
| Remove the prediction-error penalty (fully task-aware, λF = 0) (ablation) | Co-designed MPC forecaster · IoT, cell traffic, and battery data | forecast quality | Hurt |
| Use an LSTM forecaster instead of a feedforward forecaster, or vice versa (ablation) | Co-designed forecaster · IoT, cell traffic, and battery data | performance | No clear effect |
Closest papers by meaning
- Maximizing the Value of Predictions in Control: Accuracy Is Not Enough (NeurIPS 2025 · similarity 0.82)
- Neural Conformal Control for Time Series Forecasting (AAAI 2025 · similarity 0.80)
- EControl: Fast Distributed Optimization with Compression and Error Control (ICLR 2024 · similarity 0.79)
- Carbon Footprint Reduction for Sustainable Data Centers in Real-Time (AAAI 2024 · similarity 0.79)
- SOFTS: Efficient Multivariate Time Series Forecasting with Series-Core Fusion (NeurIPS 2024 · similarity 0.79)
- Expand and Compress: Exploring Tuning Principles for Continual Spatio-Temporal Graph Forecasting (ICLR 2025 · similarity 0.78)
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.
