Enhanced Tweedie model for insurance claims using CatBoost.
problem Accurately modeling aggregate claims with zero-inflated data.
method Refined Tweedie model with boosting methods in CatBoost.
result Marked improvement in model performance for insurance analytics.
We propose a new framework of CatBoost that predicts the entire conditional distribution of a univariate response variable. In particular, CatBoostLSS models all moments of a parametric distribution (i.e., mean, location, scale and shape [LSS]) instead of the conditional mean only. Choosing from a wide range of continu…
In this paper we present CatBoost, a new open-sourced gradient boosting library that successfully handles categorical features and outperforms existing publicly available implementations of gradient boosting in terms of quality on a set of popular publicly available datasets. The library has a GPU implementation of lea…
Machine learning models outperform traditional option pricing models.
problem Improving option pricing accuracy using complex models.
method Evaluation of machine learning (NN, RF, CatBoost) and traditional models (Black-Scholes, Heston) on synthetic and real data.
result Machine learning models outperform traditional models in predicting option prices.
Unified comparison of gradient boosting algorithms for insurance claims.
problem Improving predictive accuracy and computational efficiency in insurance claim prediction.
method Unified notation and comprehensive numerical study comparing 12 gradient boosting algorithms on 5 datasets.
result No trade-off between model adequacy and predictive accuracy.
Boosting algorithms improve delivery time prediction in postal services.
problem Challenges in long-term travel time prediction for postal services.
method Investigated linear regression models, tree-based ensembles (random forest, bagging, boosting), and compared their performance.
result Boosting algorithms, especially light gradient boosting and catboost, outperform other methods in accuracy and runtime efficiency.
A machine learning model improves relative valuation of municipal bonds.
problem Challenges in determining the value or relative value of municipal bonds.
method Proposes a supervised similarity framework using CatBoost algorithm to identify similar bonds based on risk profiles.
result The similarity-based method outperforms rule-based and heuristic-based methods in back-testing.
Study uses synthetic data to estimate credit risk for underbanked consumers in Istanbul.
problem Estimating credit risk for underbanked consumers lacking formal credit records.
method Created synthetic dataset, used retrieval augmented generation, trained CatBoost, LightGBM, and XGBoost models.
result Alternative financial data improves credit risk estimation, raising AUC by 13%.
CTRNNs improve blood glucose forecasting in ICU, outperforming traditional models.
problem Forecasting blood glucose in ICU with irregular measurements.
method Continuous time autoregressive recurrent neural networks (CTRNNs) using neural ODE or neural flow layers.
result CTRNNs generally outperform traditional autoregressive models in probabilistic forecasting of blood glucose.
Study predicts coastal water quality using machine learning, identifying salinity as key factor.
problem Predicting and managing coastal water quality for public health and tourism.
method Machine learning models (Catboost, Xgboost, Random Forests, Support Vector Regression, Artificial Neural Networks) trained on environmental data.
result Catboost algorithm performed best, with R² values of 0.71 and 0.68 for E. Coli and enterococci predictions.
StructureBoost improves gradient boosting for complex categorical variables efficiently.
problem Efficiently handling complex categorical variables with known structure.
method Two methods to overcome computational obstacles in SCDT enumeration for structured categorical variables.
result StructureBoost outperforms existing packages on complex categorical problems.
Enhanced binary classifier uses Urysohn's Lemma of Topology.
problem Binary classification challenges.
method Utilizes Urysohn's Lemma of Topology to construct separating functions.
result Exceptional performance in numerical experiments (95% to 100%).
StochasticRank optimizes ranking metrics efficiently and guarantees global convergence.
problem Optimizing discrete ranking metrics due to their ill-posed nature.
method Stochastic smoothing, gradient estimate, debiasing, and Stochastic Gradient Langevin Boosting.
result Global convergence and superior performance on ranking datasets.
SAINT improves neural networks for tabular data with row attention and contrastive pre-training.
problem Tabular data challenges in machine learning applications.
method SAINT combines row and column attention with contrastive self-supervised pre-training.
result SAINT outperforms previous deep learning methods and even gradient boosting methods on benchmark tasks.
The study uses ML and AI to forecast pension fund mortality, outperforming traditional methods.
problem Incorporating longevity risk into pension fund financial assessments.
method Employed actuarial learning with ML/AI techniques (regression trees, random forest, boosting, XGBoost, CatBoost, neural networks) on actuarial data.
result ML/AI algorithms outperform the Lee-Carter model in mortality forecasting for pension funds.
Machine learning models predict the behavior of negatively buoyant jets from wastewater.
problem Minimizing harmful effects of negatively buoyant jets during wastewater discharge.
method Training machine learning models (ANN, XGBoost, CatBoost, LightGBM) on OpenFOAM simulations and experimental data.
result Artificial Neural Network provided the best prediction with R2 0.98 and RMSE 0.28.
This study evaluates feature scaling across 14 datasets and 12 techniques in ML.
problem Impact of feature scaling on machine learning performance and computational costs.
method Systematic evaluation of 12 scaling techniques across 14 datasets and 16 ML algorithms.
result Wide variation in model performance due to feature scaling, especially for non-ensemble models.
Gradient boosting decision trees (GBDTs) have seen widespread adoption in academia, industry and competitive data science due to their state-of-the-art performance in many machine learning tasks. One relative downside to these models is the large number of hyper-parameters that they expose to the end-user. To maximize …
Study predicts startup outcomes like funding, patenting, IPOs using machine learning.
problem Forecasting startup success metrics like funding, patenting, IPOs.
method Developed interpretable machine learning framework, used preprocessing, class imbalance handling, and compared multiple models.
result Achieved high AUROC values for patent, funding, and exit predictions.
Stochastic Gradient Boosting (SGB) is a widely used approach to regularization of boosting models based on decision trees. It was shown that, in many cases, random sampling at each iteration can lead to better generalization performance of the model and can also decrease the learning time. Different sampling approaches…
We describe our first-place solution to the Animal Behavior Challenge (ABC 2018) on predicting gender of bird from its GPS trajectory. The task consisted in predicting the gender of shearwater based on how they navigate themselves across a big ocean. The trajectories are collected from GPS loggers attached on shearwate…
This paper proposes an embedding-based neural network for more accurate investment return prediction.
problem Accurately predicting investment returns requires understanding industry knowledge and news, as well as leveraging relevant theories.
method The approach uses embedding to encode investment IDs into low-dimensional vectors, leveraging dual branches to separate different information, and employs the swish activation function.
result The proposed embedding-based dual branch model outperforms traditional machine learning models like Xgboost, Lightgbm, and Catboost on the Ubiquant Market Prediction dataset.
Cryptocurrency patterns stable across market caps, validated by microstructure theory.
problem Stable patterns in cryptocurrency microstructure across different market caps.
method Unified CatBoost modeling pipeline with time-series cross validation, validated by backtests.
result Feature rankings and partial effects are stable across assets despite heterogeneous liquidity and volatility.
Attention augments forest for tabular data accuracy.
problem Training tabular data models with high accuracy and efficiency.
method Tree Attention Block (TAB) in differentiable forest framework.
result Attention augmented differentiable forest achieves comparable and sometimes higher accuracy than GBDT models.
Adversarial Robustness Toolbox (ART) is a Python library supporting developers and researchers in defending Machine Learning models (Deep Neural Networks, Gradient Boosted Decision Trees, Support Vector Machines, Random Forests, Logistic Regression, Gaussian Processes, Decision Trees, Scikit-learn Pipelines, etc.) agai…
XGBoost outperforms other models in predicting housing prices.
problem Accurate housing price prediction for socio-economic development.
method Employed XGBoost and other machine learning algorithms on housing price datasets.
result XGBoost outperformed other models in predicting housing prices.
XGBoost implements AFT models for survival regression.
problem Survival regression for time-to-event data.
method Loss functions for AFT models in XGBoost.
result XGBoost's AFT model improves generalization and training speed.
New algorithm improves convergence of gradient boosting trees.
problem Global convergence of Newton boosting in tabular machine learning.
method Introduces Gradient Regularized Newton Descent for GBDTs, proving linear convergence for smooth, strongly convex losses and O(k21) rate for general convex losses. result Achieves globally convergent second-order GBDT algorithm with rate matching first-order boosting.
Two new techniques improve zero-shot HPO efficiency.
problem Efficiently selecting hyperparameters for new datasets.
method Surrogate model and multi-fidelity techniques.
result Significant improvement in accuracy compared to standard methods.
A new framework separates classifier calibration and discrimination.
problem Combining reliability and resolution in probabilistic predictions.
method Manokhin Probability Matrix separates reliability and resolution using Spiegelhalter Z-statistic and AUC-ROC.
result Classifiers are categorized into four archetypes: Eagle, Bull, Sloth, and Mole.
Drift-Resilient TabPFN learns to adapt to changing data distributions.
problem Real-world data often shifts over time, degrading model performance.
method In-Context Learning with a Prior-Data Fitted Network, using structural causal models.
result Significant performance improvements across various datasets.
CCI combines Bayesian and gradient boosting to create fair, reliable credit risk scores.
problem Tackles high-stakes lending decisions with changing data distributions and fairness constraints.
method Combines Bayesian neural risk scorer and fairness-constrained gradient boosting with shift-aware fusion.
result CCI achieves best trade-off between discrimination, calibration, stability, and fairness.
Study compares various calibration methods for binary classification tasks.
problem Improving probabilistic predictions in binary classification models.
method Benchmarked 21 classifiers using 5 calibration methods on real data.
result Venn-Abers predictors and Beta calibration show the largest log-loss reductions.
Donor-aware scRNA-seq benchmarks improve classification accuracy in inflammatory bowel disease.
problem Influenza disease classification from scRNA-seq data is prone to donor-level confounding.
method Developed and evaluated three feature representations across two IBD cohorts.
result Compartment-stratified CLR composition and GatedStructuralCFN embeddings outperform linear models in classification accuracy.