用蒙特卡洛方法生成序列预测的置信区间，提升风险敏感场景下的可靠性。
A Monte Carlo Framework for Calibrated Uncertainty Estimation in Sequence Prediction
- 通过自回归神经网络模拟器采样序列，估计概率分布和置信区间。
- 在合成与真实数据上验证，修正后模型预测校准度显著提升。
- 适合需要量化不确定性的高风险应用，如自动驾驶、医疗诊断。
从图像等高维数据中进行序列的概率预测是关键挑战，尤其在风险敏感场景中，仅给出最可能序列（如语言建模）不足以满足需求。本文提出一种蒙特卡洛框架，用于估计离散序列分布的概率及置信区间。该框架采用自回归训练的神经网络作为蒙特卡洛模拟器，根据图像输入生成条件序列样本，并基于这些样本进行概率与置信区间的估算。在合成数据和真实数据上的实验表明，该框架能实现准确的判别性预测，但存在校准不足问题。为此，我们引入时间依赖正则化方法，有效提升了预测的校准性能。
原文摘要 · Abstract (English)
Probabilistic prediction of sequences from images and other high-dimensional data is a key challenge, particularly in risk-sensitive applications. In these settings, it is often desirable to quantify the uncertainty associated with the prediction (instead of just determining the most likely sequence, as in language modeling). In this paper, we propose a Monte Carlo framework to estimate probabilities and confidence intervals associated with the distribution of a discrete sequence. Our framework uses a Monte Carlo simulator, implemented as an autoregressively trained neural network, to sample sequences conditioned on an image input. We then use these samples to estimate the probabilities and confidence intervals. Experiments on synthetic and real data show that the framework produces accurate discriminative predictions, but can suffer from miscalibration. In order to address this shortcoming, we propose a time-dependent regularization method, which is shown to produce calibrated predictions.
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。