초록 방어 계획: FL Framework 논문으로의 포지셔닝
Source:
report/version6/lab-leader/v6_0418_abstract_defense_plan.md
초록 방어 계획: Claim별 증거 연결 및 실험 설계
0. 전제: 방향 재정의
v2 보고서(v6_0418_fl_baseline_critique.md)는 "선택지 B(Peak Loss 단독 논문)로 전환"을 권고했다.
그러나 사용자가 아래 초록을 변경 불가 확정 상태로 전달했다. 초록은 명시적으로 FL framework 논문으로 포지셔닝된다.
Accurate peak load forecasting is critical to power system operation, but household-level forecasting remains difficult due to load variability. Since consumption data are privacy-sensitive, federated learning (FL) offers a natural framework for collaborative modeling without centralized data collection.
We propose a peak-aware FL framework with two components: (i) a peak-weighted SmoothL1 loss that emphasizes peak-time samples during training, and (ii) a dual-path architecture that combines a shared vector-quantized (VQ) codebook for federated pattern exchange with a lightweight DLinear residual for per-household personalization. We evaluate models using PAPE (Peak Absolute Percentage Error) and HR (Hit Rate), metrics tailored to peak-region accuracy.
Across a broad set of baselines, the proposed model achieves the best peak accuracy, reducing PAPE by 9.8% relative to local-only training. Notably, our 4.6K-parameter model surpasses a 50M-parameter foundation model, showing that loss design and personalization outweigh model scale for household peak forecasting under privacy constraints.