Research map / Bandits and online learning / Online convex optimization and games
Learning Augmented Energy Minimization via Speed Scaling
Etienne Bamas, Andreas Maggiori, Lars Rohwedder, Ola Svensson
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 LAS with accurate predictions instead of best online algorithm (OA) | LAS · Synthetic, accurate predictor | competitive ratio: 1.199 -> 1.008 | Improved |
| use LAS with uninformative random predictions vs OA | LAS · Synthetic, random predictor | competitive ratio: 1.199 -> 1.239 | No clear effect |
| use LAS with misleading predictions vs OA | LAS · Synthetic, misleading predictor | competitive ratio: 1.361 -> 1.766 | Hurt |
| increase robustness parameter epsilon from 0.01 to 0.8 (ablation) | LAS · Synthetic, accurate predictor | competitive ratio: 1.008 -> 1.026 | Hurt |
| increase robustness parameter epsilon from 0.01 to 0.8 (ablation) | LAS · Synthetic, misleading predictor | competitive ratio: 1.766 -> 1.750 | Improved |
| increase robustness parameter epsilon from 0.01 to 0.8 (ablation) | LAS · Synthetic, random predictor | competitive ratio: 1.239 -> 1.203 | Improved |
| use LAS with previous-day prediction vs AVR and OA | LAS · BrightKite real data | competitive ratio: 1.36 and 1.24 -> 1.116 | Improved |
| increase epsilon from 0.01 to 0.8 on low-error days (ablation) | LAS · BrightKite, first 100 days | competitive ratio: 1.134 -> 1.146 | Hurt |
Closest papers by meaning
- Energy-Efficient Scheduling with Predictions (NeurIPS 2023 · similarity 0.89)
- Power Lines: Scaling laws for weight decay and batch size in LLM pre-training (NeurIPS 2025 · similarity 0.83)
- The ML.ENERGY Benchmark: Toward Automated Inference Energy Measurement and Optimization (NeurIPS 2025 · similarity 0.83)
- Carbon Footprint Reduction for Sustainable Data Centers in Real-Time (AAAI 2024 · similarity 0.82)
- DCcluster-Opt: Benchmarking Dynamic Multi-Objective Optimization for Geo-Distributed Data Center Workloads (NeurIPS 2025 · similarity 0.81)
- Freya PAGE: First Optimal Time Complexity for Large-Scale Nonconvex Finite-Sum Optimization with Heterogeneous Asynchronous Computations (NeurIPS 2024 · similarity 0.81)
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.
