Signal-Aware Workload Shifting Algorithms with Uncertainty-Quantified Predictors

2025-09-30

AI Summary

This paper addresses the challenge of uncertainty in external signals—such as grid carbon intensity and electricity prices—in online workload scheduling. To tackle this, we propose UQ-Advice, a novel algorithm that integrates an Uncertainty Quantification (UQ) predictor with a learning-enhanced online decision framework, systematically incorporating prediction uncertainty as an explicit input to scheduling decisions for the first time. Our key innovations include a “decision uncertainty score” mechanism and a UQ-robustness metric, enabling credibility-aware optimization of scheduling timing. Extensive experiments on real-world datasets demonstrate that UQ-Advice significantly outperforms baseline methods ignoring uncertainty, under both carbon-intensity- and price-driven scheduling scenarios. Results confirm that explicitly leveraging uncertainty information improves both the reliability and cost-efficiency of green workload scheduling.

Technology Category

• Planning, Routing, and Scheduling: Scheduling under Uncertainty
• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Sequential Decision Making

Application Category

• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

A wide range of sustainability and grid-integration strategies depend on workload shifting, which aligns the timing of energy consumption with external signals such as grid curtailment events, carbon intensity, or time-of-use electricity prices. The main challenge lies in the online nature of the problem: operators must make real-time decisions (e.g., whether to consume energy now) without knowledge of the future. While forecasts of signal values are typically available, prior work on learning-augmented online algorithms has relied almost exclusively on simple point forecasts. In parallel, the forecasting research has made significant progress in uncertainty quantification (UQ), which provides richer and more fine-grained predictive information. In this paper, we study how online workload shifting can leverage UQ predictors to improve decision-making. We introduce $ exttt{UQ-Advice}$, a learning-augmented algorithm that systematically integrates UQ forecasts through a $ extit{decision uncertainty score}$ that measures how forecast uncertainty affects optimal future decisions. By introducing $ extit{UQ-robustness}$, a new metric that characterizes how performance degrades with forecast uncertainty, we establish theoretical performance guarantees for $ exttt{UQ-Advice}$. Finally, using trace-driven experiments on carbon intensity and electricity price data, we demonstrate that $ exttt{UQ-Advice}$ consistently outperforms robust baselines and existing learning-augmented methods that ignore uncertainty.

Problem

• Online workload shifting with uncertainty-quantified predictors
• Improving real-time energy decisions using forecast uncertainty
• Leveraging uncertainty quantification for grid integration strategies

Innovation

• Uses uncertainty-quantified predictors for workload shifting
• Introduces decision uncertainty score for forecast integration
• Establishes UQ-robustness metric with theoretical guarantees