Modular Deep Learning Mechanisms for Auditable Next-Day Wildfire Spread Prediction
研究通过引入风向和坡度条件注意力偏差、物理特征检索增强输出校正及火情条件双流门控三种模块化增强方法，提升次日野火蔓延预测模型的可审计性和预测准确性。
0 citationsRead paper
Selected work
Representative Papers
A family of spectral conjugate gradient algorithms derived by least-squares approximations based on a modified quasi--Newton update with application to a revised robust binary classification model
本文提出了一种基于改进拟牛顿更新的谱共轭梯度算法，用于解决无约束优化问题，并应用于鲁棒二分类模型中以提高准确性和训练效率。
0 citationsRead paper
MURAL: Multimodal Uncertainty-aware Recommendation via Adaptive edge Learning
本文提出MURAL框架，通过自适应边学习和不确定性感知融合解决推荐系统中的结构僵化和语义脆弱问题。
0 citationsRead paper
BLADE: Bilevel Low-rank Augmented-Lagrangian Erasure for LLM Unlearning
BLADE通过双层低秩增广拉格朗日机制解决大模型遗忘中的鲁棒性问题，提高模型在多次遗忘操作后的稳定性和性能。
0 citationsRead paper
Constructing and extending $n$ = 1 Bayesian confidence intervals for location parameters in location-scale families
This study addresses the long-standing challenge of constructing nontrivial, efficient confidence intervals for the location parameter of a location-scale family when only a single observation is available. The authors propose two Bayesian approaches: first, deriving priors that yield asymptotically efficient intervals at high confidence levels; second, integrating classical t-intervals with prior information to form an enhanced t-interval based on Bayes factor testing. The work establishes, for the first time, a systematic Bayesian mechanism for generating single-sample confidence intervals and demonstrates an equivalence between Bayesian credible intervals and frequentist confidence intervals. The methodology extends to any continuous symmetric location-scale family. Theoretical results show that the proposed intervals are asymptotically efficient when \( n = 1 \), and for \( n \geq 2 \), the enhanced t-interval achieves smaller expected squared width over parts of the parameter space, with practical utility validated on interstellar object velocity data.
0 citationsRead paper
Recent publications
Latest Papers
Modular Deep Learning Mechanisms for Auditable Next-Day Wildfire Spread Prediction
研究通过引入风向和坡度条件注意力偏差、物理特征检索增强输出校正及火情条件双流门控三种模块化增强方法，提升次日野火蔓延预测模型的可审计性和预测准确性。
0 citationsRead paper
A family of spectral conjugate gradient algorithms derived by least-squares approximations based on a modified quasi--Newton update with application to a revised robust binary classification model
本文提出了一种基于改进拟牛顿更新的谱共轭梯度算法，用于解决无约束优化问题，并应用于鲁棒二分类模型中以提高准确性和训练效率。
0 citationsRead paper
MURAL: Multimodal Uncertainty-aware Recommendation via Adaptive edge Learning
本文提出MURAL框架，通过自适应边学习和不确定性感知融合解决推荐系统中的结构僵化和语义脆弱问题。
0 citationsRead paper
BLADE: Bilevel Low-rank Augmented-Lagrangian Erasure for LLM Unlearning
BLADE通过双层低秩增广拉格朗日机制解决大模型遗忘中的鲁棒性问题，提高模型在多次遗忘操作后的稳定性和性能。
0 citationsRead paper
Constructing and extending $n$ = 1 Bayesian confidence intervals for location parameters in location-scale families
This study addresses the long-standing challenge of constructing nontrivial, efficient confidence intervals for the location parameter of a location-scale family when only a single observation is available. The authors propose two Bayesian approaches: first, deriving priors that yield asymptotically efficient intervals at high confidence levels; second, integrating classical t-intervals with prior information to form an enhanced t-interval based on Bayes factor testing. The work establishes, for the first time, a systematic Bayesian mechanism for generating single-sample confidence intervals and demonstrates an equivalence between Bayesian credible intervals and frequentist confidence intervals. The methodology extends to any continuous symmetric location-scale family. Theoretical results show that the proposed intervals are asymptotically efficient when \( n = 1 \), and for \( n \geq 2 \), the enhanced t-interval achieves smaller expected squared width over parts of the parameter space, with practical utility validated on interstellar object velocity data.
0 citationsRead paper