Algorithms最新文献

Explainable Machine Learning (XML) in high-stakes domains demands reproducible methods to aggregate feature importance across multiple models applied to the same structured dataset. We propose the Weighted Importance Score and Frequency Count (WISFC) framework, which combines importance magnitude and consistency by aggregating ranked outputs from diverse explainers. WISFC assigns a weighted score to each feature based on its rank and frequency across model-explainer pairs, providing a robust ensemble feature-importance ranking. Unlike simple consensus voting or ranking heuristics that are insufficient for capturing complex relationships among different explainer outputs, WISFC offers a more principled approach to reconciling and aggregating this information. By aggregating many "weak signals" from brute-force modeling runs, WISFC can surface a stronger consensus on which variables matter most. The framework is designed to be reproducible and generalizable, capable of taking important outputs from any set of machine-learning models and producing an aggregated ranking highlighting consistently important features. This approach acknowledges that any single model is a simplification of complex, multidimensional phenomena; using multiple diverse models, each optimized from a different perspective, WISFC systematically captures different facets of the problem space to create a more structured and comprehensive view. As a consequence, this study offers a useful strategy for researchers and practitioners who seek innovative ways of exploring complex systems, not by discovering entirely new variables but by introducing a novel mindset for systematically combining multiple modeling perspectives.

Integer linear programs (ILPs) and mixed integer programs (MIPs) often have multiple distinct optimal solutions, yet the widely used Gurobi optimization solver returns certain solutions at disproportionately high frequencies. This behavior is disadvantageous, as, in fields such as biomedicine, the identification and analysis of distinct optima yields valuable domain-specific insights that inform future research directions. In the present work, we introduce MORSE (Multiple Optima via Random Sampling and careful choice of the parameter Epsilon), a randomized, parallelizable algorithm to efficiently generate multiple optima for ILPs. MORSE maps multiplicative perturbations to the coefficients in an instance's objective function, generating a modified instance that retains an optimum of the original problem. We formalize and prove the above claim in some practical conditions. Furthermore, we prove that for 0/1 selection problems, MORSE finds each distinct optimum with equal probability. We evaluate MORSE using two measures; the number of distinct optima found in $r$ independent runs, and the diversity of the list (with repetitions) of solutions by average pairwise Hamming distance and Shannon entropy. Using these metrics, we provide empirical results demonstrating that MORSE outperforms the Gurobi method and unweighted variations of the MORSE method on a set of 20 Mixed Integer Programming Library (MIPLIB) instances and on a combinatorial optimization problem in cancer genomics.

The purpose of anomaly detection is to identify special data points or patterns that significantly deviate from the expected or typical behavior of the majority of the data, and it has a wide range of applications across various domains. Most existing statistical and machine learning-based anomaly detection algorithms face challenges when applied to high-dimensional data. For instance, the unconstrained least-squares importance fitting (uLSIF) method, a state-of-the-art anomaly detection approach, encounters the unboundedness problem under certain conditions. In this study, we propose a scaled Bregman divergence-based anomaly detection algorithm using both least absolute deviation and least-squares loss for parameter learning. This new algorithm effectively addresses the unboundedness problem, making it particularly suitable for high-dimensional data. The proposed technique was evaluated on both synthetic and real-world high-dimensional time series datasets, demonstrating its effectiveness in detecting anomalies. Its performance was also compared to other density ratio estimation-based anomaly detection methods.