A private Wilcoxon test reduces data needed for paired data analysis.
problem Performing Wilcoxon signed-rank tests on private data.
method Developed a differentially private method for computing the Wilcoxon signed-rank test.
result Our private test requires less data to achieve the same statistical power.
The paper explores how to reduce classification tasks to optimization problems in Euclidean space.
problem Understanding the minimum dimension needed for reducing classification tasks to optimization problems.
method Developed a generalization of the Borsuk-Ulam Theorem to analyze the expressivity of reductions.
result The minimum Euclidean dimension required can be exponentially larger than the VC dimension, even for slightly non-trivial reductions.
A new algorithm enhances minority class representation in imbalanced datasets.
problem Improving classification performance on imbalanced datasets.
method PO-QG algorithm using Proxima-Orion neighbors and q-Gaussian weighting.
result The PO-QG algorithm improves overall classification performance.
A new method QMS22 for semi-supervised anomaly detection outperforms existing methods.
problem Semi-supervised anomaly detection in datasets with overlapping normal and outlier samples.
method QMS22, a classifier that solves a multi-class classification problem involving both training and test sets.
result QMS22 significantly outperforms ISOF and ocSVM in anomaly detection.
Study examines complexity of opinions in online discussions.
problem Understanding the complexity and similarity of opinions in online discussions.
method Developed a modeling framework to construct latent representations of opinions, connecting them to voting models and sign-rank of a matrix.
result Found that unidimensional models often fail to accurately represent online discussions, revealing insights into human judgements and opinions.
A new framework for paired-sample testing in high-dimensional data.
problem Challenges in standard paired-sample testing for high-dimensional data.
method Develops a two-step testing procedure using scoring functions and Wilcoxon signed-rank test.
result Substantial performance gains in testing accuracy compared to traditional methods.
Study on lower bounds for deep neural networks using ReLU gates.
problem Understanding the role of depth in neural networks computing Boolean functions.
method Use of random restrictions and sign-rank methods to show lower bounds.
result Exponential lower bounds for ReLU circuits ending in a LTF gate.
A new family of nonparametric statistics, the r-statistics, is introduced. It consists of counting the number of records of the cumulative sum of the sample. The single-sample r-statistic is almost as powerful as Student's t-statistic for Gaussian and uniformly distributed variables, and more powerful than the sign and…
SBSS uses similarity to split data for better classifier training.
problem Training better classifiers with realistic performance estimation.
method SBSS uses both input and output space information to split data using similarity functions.
result SBSS outperformed ordinary stratified 10-fold cross-validation in 75% of scenarios.
Neural network model improves sepsis detection and prediction in ICU patients.
problem Improving sepsis detection and prediction in ICU patients.
method Rule-based and machine learning models, neural network ensemble model.
result Neural network model achieves highest AUC in detecting and predicting sepsis, severe sepsis, and septic shock.
The statistical comparison of multiple algorithms over multiple data sets is fundamental in machine learning. This is typically carried out by the Friedman test. When the Friedman test rejects the null hypothesis, multiple comparisons are carried out to establish which are the significant differences among algorithms. …
LSTM-FCN and ALSTM-FCN improve time series classification performance.
problem Improving time series classification performance.
method Ablation tests on LSTM-FCN and ALSTM-FCN, comparing z-normalizing techniques, dimension shuffle impact, and GRU replacement.
result LSTM and FCN blocks perform better together, and z-normalizing the whole dataset is more effective.
RL-Exec uses reinforcement learning to optimize BTC-USD liquidation, outperforming traditional methods.
problem Optimizing liquidation strategies on BTC-USD limit-order books with transient impact and latency.
method PPO agent trained on historical BTC-USD limit-order book replays, incorporating impact resilience and fees.
result RL-Exec significantly outperforms TWAP and a VWAP-like baseline on BTC-USD liquidation, with performance improving with longer execution horizons.
Bayesian model compares ML algorithms on various datasets.
problem Comparing multiple machine learning algorithms across multiple datasets.
method Bayesian Bradley-Terry model, defining regions of practical equivalence (ROPE).
result Allows nuanced statements and ROPE definitions for algorithm comparison.
XGBoost model improves business risk classification with feature selection and Bayesian hyper-parameter optimization.
problem Improving business risk classification models using advanced machine learning techniques.
method XGBoost, feature selection (FS), Bayesian hyper-parameter optimization (TPE and RS), 10-fold cross-validation.
result Bayesian TPE optimization outperforms random search and achieves higher accuracy and AUC, recall, and F1 score.
Develops a new nonparametric trace regression model for high-dimensional data.
problem Violation of known functional form and global low-rank structure assumptions in trace regression.
method Structured sign series representations for nonparametric trace regression models.
result Establishes excess risk bounds and sample complexities for the proposed model.
MSB framework improves survival prediction in immunotherapy patients with missing data.
problem High dimensionality and blockwise missingness in multimodal clinical data.
method MSB is a late-fusion framework that independently models modality-specific features before aggregating predictions via cross-validated stacking.
result MSB outperformed baseline algorithms in predicting progression-free survival in lung cancer patients.
Deep learning network matches radiologists in breast cancer segmentation.
problem Automating radiologist-level cancer segmentation from breast MRI.
method 3D U-Net architecture trained on 382,290 breast scans, compared to 255,500 benign cases.
result Network performance matched radiologists' on 2D segmentation of breast cancers.
Exchange Traded Funds (ETFs) have been gaining increasing popularity in the investment community as is evidenced by the high growth both in the number of ETFs and their net assets since 2000. As ETFs are in nature similar to index mutual funds, in this paper we examined if this growing demand for ETFs can be explained …
Proposes LSTM algorithm for robust Alzheimer's disease progression modeling with missing data.
problem Challenges in modeling disease progression using incomplete longitudinal data.
method Utilizes Long Short-Term Memory (LSTM) networks for Alzheimer's disease progression modeling with a generalized training rule for handling missing data.
result Achieves significantly lower mean absolute error (MAE) than alternatives with p < 0.05.