Lasso
PulseAugur coverage of Lasso — every cluster mentioning Lasso across labs, papers, and developer communities, ranked by signal.
2 天有情绪数据
-
Radiogenomic models predict glioblastoma immune signatures
Researchers have developed radiogenomic models capable of non-invasively predicting a specific immune cell signature in glioblastoma. These models utilize radiomic features extracted from MRI scans and transcriptomic da…
-
Researchers analyze $\ell_1$ implicit bias in $\ell_2$-boosting for benign overfitting
Researchers have analyzed the high-dimensional risk of $\ell_2$-Boosting in the context of $\ell_1$ implicit bias, identifying a logarithmic rate of excess variance decay under a pure-noise model. This phenomenon, where…
-
TinyBayes 使边缘设备上的实时作物病害检测成为可能
研究人员开发了 TinyBayes，一个用于边缘设备实时图像分类的新颖框架，专门用于检测可可作物的病害。该系统集成了闭式贝叶斯分类器和移动级计算机视觉管道，模型总大小不到 9.5 MB。TinyBayes 在 Amini 可可污染挑战数据集上实现了 78.7% 的准确率，并且可以在 CPU 上在 150 毫秒内完成推理。
-
新方法通过集成模型和最坏情况分布分析增强鲁棒优化
研究人员开发了用于分布鲁棒优化（一种考虑数据分布不确定性的技术）的新方法。一种方法是集成分布鲁棒贝叶斯优化（Ensemble Distributionally Robust Bayesian Optimization），它使用模型集成来提高鲁棒性并实现理论上的次线性遗憾界限。另一篇论文介绍了分布鲁棒多目标优化（DR-MOO），其算法在最坏情况分布下最小化目标，从而提高了样本复杂度。此外，还提出了一个用于分布鲁棒学习的框架，以优化一阶方…
-
Bayesian methods outperform classical sparse regression in prediction and uncertainty
A new benchmark study evaluated six sparse regression methods, comparing classical approaches like Lasso with Bayesian techniques such as Horseshoe and Spike-and-Slab. The research found that Bayesian methods generally …
-
New 2D Stability Selection method improves feature selection robustness
Researchers have developed a new method called "2D Stability Selection" to improve feature selection in high-dimensional regression. This technique addresses instability arising from both sampling variability and measur…
-
Researchers propose new method to stabilize private LASSO under heterogeneous covariates
Researchers have developed a new method to stabilize the LASSO algorithm when dealing with heterogeneous covariate scales under differential privacy constraints. Their approach, termed Gram-based anisotropic objective p…
-
GRASP框架通过强大的特征选择增强医学预测能力
研究人员开发了GRASP，一个用于医学预测任务的特征选择新框架。GRASP结合了Shapley值归因和组$L_{21}$正则化，以识别紧凑且可解释的特征集。该方法旨在通过提供更稳定、冗余更少的特征选择，同时保持或提高预测准确性，来改进现有的LASSO等技术。
-
SHIFT estimator improves robust double machine learning for heavy-tailed data
Researchers have developed SHIFT, a new robust estimator for Double Machine Learning (DML) pipelines designed to handle heavy-tailed data contamination. SHIFT combines cross-fit nuisance orthogonalization with a kernel-…
-
New ensemble learning framework predicts groundwater heavy metal pollution
Researchers have developed a new ensemble machine learning framework to predict groundwater heavy metal pollution in the Densu Basin. The study integrated response transformations, including a Gaussian copula, with six …
-
AI approach enhances variable selection in linear regression models
Researchers have developed a novel Artificial Intelligence approach for variable selection in linear regression models. This method utilizes an Artificial Neural Network (ANN) trained to assess variable significance bas…
-
New FEA method speeds up entropic measure computation for ML
Researchers have developed Fast Entropic Approximations (FEA), a new method for approximating entropic measures like Shannon entropy and Kullback-Leibler divergence. These approximations are non-singular, property-prese…
-
New online algorithm enhances high-dimensional probabilistic electricity price forecasting
Researchers have developed an online algorithm for multivariate distributional regression to forecast electricity prices, addressing the underexplored multivariate nature of day-ahead prices. This method efficiently mod…
-
New algorithm enables efficient online estimation of distributional models
Researchers have introduced a new methodology for online estimation of regularized, linear distributional models, designed to handle large-scale streaming data. This approach combines advancements in online LASSO model …
-
Hugging Face paper introduces SimpleTES framework for scaling LLM-driven scientific discovery
Researchers have introduced a framework called Simple Test-time Evaluation-driven Scaling (SimpleTES) to enhance the scalability of language model-driven scientific discovery. This method strategically combines parallel…
-
Machine learning model homotopy explores coefficient sign changes
The concept of model homotopy, applying topological ideas to machine learning, suggests that a single model may not fully capture a modeling situation. Instead, a trajectory of fits, parameterized continuously by weight…