Papers

决策聚焦学习新框架:以 CVaR 最小化遗憾尾部风险

Risk-Averse Decision-Focused Learning: Minimizing CVaR of Regret

Why we picked it

做优化+预测结合方向的人可以看这篇:不改推理流程,只靠一个替代损失就把 SPO+ 这类方法变成风险敏感版本,还带泛化界。四类问题的实验验证了在高风险场景下尾部遗憾明显更低,比在推理时跑鲁棒优化省计算。

arXiv 论文提出一个风险规避的决策聚焦学习(DFL)框架,不再像多数现有方法那样只最小化期望遗憾,而是直接最小化遗憾的 CVaR(条件风险价值),以覆盖高风险尾部场景。

Body

方法从 CVaR 的对偶表示出发,加入熵正则化,推导出遗憾 CVaR 的光滑替代损失,推理阶段仍保留点预测和确定性优化,避免先前风险敏感方法需要显式建模不确定性并在推理时求解随机或鲁棒优化问题带来的计算开销。梯度在 minibatch 内形成实例依赖的重加权,强调高风险样本,因此可以把 SPO+ 等现有替代损失 DFL 方法直接改造成风险敏感版本,且基于 SPO+ 实例化时对真实遗憾的 CVaR 存在泛化界。

实验覆盖 knapsack、组合投资、最短路径和能耗调度四类问题,显示当不确定参数呈现异质尾部行为时,尾部遗憾获得大幅下降。

Original · arXiv cs.AI

Risk-Averse Decision-Focused Learning: Minimizing CVaR of Regret

Many combinatorial optimization (CO) problems involve parameters that are uncertain when the problem must be solved. In such settings, the parameters have to be predicted from correlated contextual information before a solution can be computed. \emph{Decision-focused learning} (DFL) addresses this by training the predictive model to produce predictions that optimize downstream decision quality. Most existing DFL methods are risk-neutral and minimize the expected regret. In high-stakes applications this can lead to undesirable outcomes, because decision-makers must also account for the tail risk of decision quality. Prior risk-sensitive approaches typically model uncertainty explicitly and solve a stochastic or robust optimization problem at inference time, which increases computational complexity. We instead propose a DFL framework that achieves risk-sensitive behavior while retaining point predictions and deterministic optimization at inference time. Starting from the dual representation of the Conditional Value-at-Risk (CVaR) and adding entropy regularization, we derive a smooth surrogate for the CVaR of regret. Its gradient induces an instance-dependent reweighting within each minibatch that emphasizes high-risk outcomes, so the approach can turn existing surrogate-loss DFL methods, including the SPO+ loss, into risk-sensitive variants. When instantiated with SPO+, the resulting objective admits a generalization bound on the CVaR of the true regret. Experiments on knapsack, portfolio, shortest-path, and energy-aware scheduling problems show substantial reductions in tail regret when the uncertain parameters exhibit heterogeneous tail behavior.

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Papers" desk. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include CVaR, SPO+, combinatorial optimization.

Source attribution: curated from arXiv cs.AI (@Jayanta Mandi, Senne Berden, Victor Bucarey, Concepcion Dominguez, Tias Guns).