arXiv research

A new algorithm speeds up rerandomization for better experiment balance.

New approach finds analytic interpretation of algebraic invariants for balanced metrics.

Unified formula for training dynamics of linear networks combining lazy and balanced regimes.

I show the equivalence between a model of financial contagion and the threshold model of global cascades proposed by Watts (2002). The model financial network comprises banks that hold risky external assets as well as interbank assets. It is shown that a simple threshold model can replicate the size and the frequency o…

Adaptive algorithm for outlier detection by balancing arm exploration and threshold estimation.

Detects crypto pump-and-dump schemes with a thresholding-based model.

New method optimizes tail dependence coefficient estimation.

Synthetic data augmentation can improve imbalanced classification metrics.

In high-dimensional classification settings, we wish to seek a balance between high power and ensuring control over a desired loss function. In many settings, the points most likely to be misclassified are those who lie near the decision boundary of the given classification method. Often, these uninformative points sho…

Model shows worldwide trade crises can be localized or global, depending on trade balance.

"Sparse" neural networks, in which relatively few neurons or connections are active, are common in both machine learning and neuroscience. Whereas in machine learning, "sparsity" is related to a penalty term that leads to some connecting weights becoming small or zero, in biological brains, sparsity is often created wh…

Study membership inference under skewed priors and adaptive thresholds, improving attack accuracy.

New method converts conventional ANNs to SNNs with minimal loss and efficiency.

Deep ROC analysis improves model selection and interpretation in medical and AI applications.

Noise in linear networks minimizes sharpness and leads to shrinkage-thresholding.

MTSSL optimizes threshold τ for better semi-supervised learning performance.

We construct the Google matrices of bitcoin transactions for all year quarters during the period of January 11, 2009 till April 10, 2013. During the last quarters the network size contains about 6 million users (nodes) with about 150 million transactions. From PageRank and CheiRank probabilities, analogous to trade imp…

Development of stock networks is an important approach to explore the relationship between different stocks in the era of big-data. Although a number of methods have been designed to construct the stock correlation networks, it is still a challenge to balance the selection of prominent correlations and connectivity of …

The personal income distribution (PID) above the Pareto threshold is studied and modeled. A microeconomic model is proposed to simulate the PID and its evolution below and above the Pareto income threshold. The model balances processes of income production and dissipation for any person above 15 years of age. The model…

Class imbalance presents a major hurdle in the application of data mining methods. A common practice to deal with it is to create ensembles of classifiers that learn from resampled balanced data. For example, bagged decision trees combined with random undersampling (RUS) or the synthetic minority oversampling technique…

A new method optimizes material discovery by balancing exploration and exploitation.

CTI produces efficient prediction intervals with guaranteed coverage.

New RL method learns K-step lookahead Q-functions for fixed-horizon MDPs.

Meta-gradient D4PG optimizes performance and constraint adherence in RL.

Optimal rank-adaptive matrix estimation from linear measurements.

Novel active learning framework using sparse approximation for efficient model training.

Optimal control of reserve assets for stablecoins to maintain peg stability.

New insights into balancing reward and fairness in stochastic MAB.

Equity-Directed Bootstrapping improves model performance across groups in imbalanced datasets.

HARFE approximates sparse additive functions using random features and ridge regression.

Paper improves deep point cloud compression techniques.

A federated model learns shared archetypes from heterogeneous clients in continual learning.

Consistent model selection for spiked Wigner model via AIC-type criteria.

C-$t^3$VAE improves class representation in long-tailed generative models.

Recurrent neural networks and sequence to sequence models require a predetermined length for prediction output length. Our model addresses this by allowing the network to predict a variable length output in inference. A new loss function with a tailored gradient computation is developed that trades off prediction accur…

New insights into how large learning rates affect transformer training dynamics.

Neural networks improve VaR estimation accuracy and robustness.

Study exact community recovery in noisy SBM with limited queries.

SA-BCP combines long-term and local evidence for efficient, adaptive online prediction.

This paper finds the noise threshold for learning Gaussian mixture models equals channel capacity.

Study evaluates three class imbalance techniques across diverse datasets.

Let $\{α\}$ and $\{β\}$ be nef cohomology classes of bidegree $(1,\,1)$ on a compact $n$-dimensional Kähler manifold $X$ such that the difference of intersection numbers $\{α\}^n - n\,\{α\}^{n-1}.\,\{β\}$ is positive. We solve in a number of special but rather inclusive cases the quantitative part of Demailly's Transce…

A framework for multi-label sentiment analysis in 100 languages with dynamic weighting.

The intrinsic error tolerance of neural network (NN) makes approximate computing a promising technique to improve the energy efficiency of NN inference. Conventional approximate computing focuses on balancing the efficiency-accuracy trade-off for existing pre-trained networks, which can lead to suboptimal solutions. In…

Iterative thresholding algorithms seek to optimize a differentiable objective function over a sparsity or rank constraint by alternating between gradient steps that reduce the objective, and thresholding steps that enforce the constraint. This work examines the choice of the thresholding operator, and asks whether it i…

This work improves privacy in federated combinatorial bandits by balancing regret and privacy.

Investigates timing and asset allocation for life insurance in uncertain financial planning.

Study evaluates machine learning methods for large-scale network reliability, revealing ANN's and PR's performance.