arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2206.10697cs.LG
收藏

数据集迁移下的性能预测

Performance Prediction Under Dataset Shift

Dataiku(Dataiku公司)

Simona Maggio, Victor Bouvier, Léo Dreyfus-Schmidt

摘要:

部署在生产环境中的ML模型通常必须面对未知的领域变化,这与其训练设置有着根本的不同。性能预测模型执行测量这些变化对模型性能的影响的关键任务。我们研究了各种性能预测模型的泛化能力,以新的领域通过学习生成的合成扰动。在十个表格数据集的基准上进行的经验验证表明,基于最先进的移位检测指标的模型不足以概括到看不见的领域,而错误预测器在移位下的性能预测方面带来了一致的改进。我们还提出了一个自然和毫不费力的不确定性估计的预测精度,确保可靠的使用性能预测。我们的实现可以在https://github.com/dataiku-research/performance_prediction_under_shift上获得。

英文摘要:

ML models deployed in production often have to face unknown domain changes, fundamentally different from their training settings. Performance prediction models carry out the crucial task of measuring the impact of these changes on model performance. We study the generalization capabilities of various performance prediction models to new domains by learning on generated synthetic perturbations. Empirical validation on a benchmark of ten tabular datasets shows that models based upon state-of-the-art shift detection metrics are not expressive enough to generalize to unseen domains, while Error Predictors bring a consistent improvement in performance prediction under shift. We additionally propose a natural and effortless uncertainty estimation of the predicted accuracy that ensures reliable use of performance predictors. Our implementation is available at https: //github.com/dataiku-research/performance_prediction_under_shift.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑