利用任务间弱单调性，少样本学习中高效筛选模型
Hedging on the Frontier: Learning New Tasks with Few Samples
- 基于弱单调性约束筛选相关任务的模型
- 在少样本新任务上显著提升性能，优于传统方法
- 适合资源有限场景下的快速模型选择与迁移
当学习者面对少量样本的新任务时，需利用可用的附加信息。实践中，这通常体现为公共基准上相关任务的模型评估结果。关键问题在于如何建模任务相关性，使其既符合实际，又能保证基准评估带来可证明的性能提升。我们观察到，弱单调性常近似成立：若一个模型在多个基准上优于另一模型，则它也更可能在新任务上表现更好。本文研究了在（近似）弱单调性假设下的学习统计复杂度，将其应用于迁移学习和模型选择聚合两种范式。结果表明，不仅可基于单调性剪枝模型类，还能通过在前沿上进行对冲，进一步适应已有权衡关系的几何结构。
原文摘要 · Abstract (English)
When a learner faces a new task with few samples, it must leverage any available side information. In practice, this often comes in the form of model evaluations on related tasks in public benchmarks. A key question then is how to model task relatedness such that it is both realistic and the benchmark evaluations lead to provable gains. Empirically, we observe that weak monotonicity is often approximately satisfied: if a model dominates another on many benchmarks, it also tends to outperform on the new task. We explore the statistical complexity of learning under (approximate) weak monotonicity, leveraging it within two learning paradigms: transfer learning and model selection aggregation. We show that not only can we prune the model class based on monotonicity, but we can also further adapt to the geometry of the available trade-offs by hedging on the frontier.
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。