paper-with-me

Papers

Diachronic Sample Integration: Robust Tail-Risk Estimation with Generative Models

2026-07-12 · Shuning Zhao, Patrick Wong, Leran Zhang, Xiaolin Hu arxiv

Deep generative models are increasingly used as simulators for downstream decision-making under data scarcity, but in risk-sensitive applications their usefulness depends on rare adverse scenarios rather than typical samples. Standard generative objectives prioritize bulk distributional fidelity, leaving low-probability tails vulnerable to localized optimization noise and making tail-dependent functionals unstable under finite simulation budgets. We introduce Diachronic Sample Integration (DSI), a test-time inference framework that ensembles generated samples across checkpoints from a stochastic training trajectory. DSI targets a checkpoint-mixture distribution that averages checkpoint-specific tail fluctuations rather than relying on a single brittle endpoint. We formalize this mechanism through a finite-budget bias-variance theory. Empirically, across multivariate synthetic processes and high-frequency trading data, DSI substantially reduces tail-estimation error compared to single-checkpoint baselines under fixed simulation budgets, outperforming standard diffusion and state-of-the-art tail-aware baselines without modifying the generative objective.

📄 PDF Abstract BibTeX arXiv:2607.10810

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

The Privacy Price of Tail-Risk Learning: Effective Tail Sample Size in Differentially Private CVaR Optimization

2026-05-15 · El Mustapha Mansouri arxiv

Differential privacy changes the effective sample size governing CVaR learning. For tail mass $τ$, the privacy-relevant sample size is not $n$, but $nτ$; equivalently, the effective private tail sample size is $εnτ$. Pri…

Estimation of Spectral Risk Measures

2019-12-22 · Ajay Kumar Pandey, Prashanth L. A., Sanjay P. Bhat

We consider the problem of estimating a spectral risk measure (SRM) from i.i.d. samples, and propose a novel method that is based on numerical integration. We show that our SRM estimate concentrates exponentially, when t…

Numerical Integration

Estimating and backtesting risk under heavy tails

2022-01-25 · Marcin Pitera, Thorsten Schmidt

While the estimation of risk is an important question in the daily business of banking and insurance, many existing plug-in estimation procedures suffer from an unnecessary bias. This often leads to the underestimation o…

Time SeriesTime Series Analysis

Estimating and backtesting risk under heavy tails

2020-10-20 · Marcin Pitera, Thorsten Schmidt

While the {estimation} of risk is an important question in the daily business of banking and insurance, many existing plug-in estimation procedures suffer from an unnecessary bias. This often leads to the underestimation…

Time SeriesTime Series Analysis

Automatic Annotation and Manual Evaluation of the Diachronic German Corpus T\"uBa-D/DC

2012-05-01 · LREC 2012 5 · Erhard Hinrichs, Thomas Zastrow

This paper presents the Tu{\`I}ˆbingen Baumbank des Deutschen Diachron (Tu{\`I}ˆBa-D/DC), a linguistically annotated corpus of selected diachronic materials from the German Gutenberg Project. It was automatically annotat…