对比推荐系统中集成方法的精度与能耗，发现精度提升常伴随巨大能耗增加。
Ensembles at Any Cost? Accuracy-Energy Trade-offs in Recommender Systems
- 通过控制实验对比四种集成策略与单模型在精度和能耗上的表现。
- 集成使精度提升0.3%至5.7%，但能耗增加19%至2549%。
- 选择性集成比全量平均更节能，适合关注能效的部署场景。
集成方法常用于推荐系统以提升精度，但多数研究仍以精度和鲁棒性为主导，忽视能源效率。本文在两个流水线（Surprise显式评分预测、LensKit隐式反馈排序）中开展93组受控实验，评估四个数据集（MovieLens 100K、MovieLens 1M、ModCloth、Anime）上四种集成策略（平均、加权、堆叠或排名融合、最佳模型组合）相对于基线和优化单模型的表现。全系统能耗通过智能插座使用EMERS测量并转换为二氧化碳当量。结果表明：集成使精度提升0.3%至5.7%，但能耗增加19%至2549%。在MovieLens 1M上，最佳模型集成使RMSE降低0.96%，能耗增加18.8%；在MovieLens 100K上，平均集成使NDCG@10提升5.7%，能耗增加103%；在Anime数据集上，Surprise最佳模型集成使RMSE下降1.2%，但能耗飙升至2005%（0.21 vs. 0.01 Wh），CO2排放从2.6增至53.8 mg，LensKit集成因内存限制失败。总体而言，选择性集成优于全量平均。
原文摘要 · Abstract (English)
Ensemble methods are frequently used in recommender systems to improve accuracy by combining multiple models. Recent work reports sizable performance gains, but most studies still optimize primarily for accuracy and robustness rather than for energy efficiency. This paper measures accuracy energy trade offs of ensemble techniques relative to strong single models. We run 93 controlled experiments in two pipelines: 1. explicit rating prediction with Surprise (RMSE) and 2. implicit feedback ranking with LensKit (NDCG@10). We evaluate four datasets ranging from 100,000 to 7.8 million interactions (MovieLens 100K, MovieLens 1M, ModCloth, Anime). We compare four ensemble strategies (Average, Weighted, Stacking or Rank Fusion, Top Performers) against baselines and optimized single models. Whole system energy is measured with EMERS using a smart plug and converted to CO2 equivalents. Across settings, ensembles improve accuracy by 0.3% to 5.7% while increasing energy by 19% to 2,549%. On MovieLens 1M, a Top Performers ensemble improves RMSE by 0.96% at an 18.8% energy overhead over SVD++. On MovieLens 100K, an averaging ensemble improves NDCG@10 by 5.7% with 103% additional energy. On Anime, a Surprise Top Performers ensemble improves RMSE by 1.2% but consumes 2,005% more energy (0.21 vs. 0.01 Wh), increasing emissions from 2.6 to 53.8 mg CO2 equivalents, and LensKit ensembles fail due to memory limits. Overall, selective ensembles are more energy efficient than exhaustive averaging,
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。