Published inTowards AIA Novel and Practical Meta‑Booster for Supervised LearningA Stacking‑Enhanced Margin‑Space Framework for Dynamic, Loss‑Driven Ensemble Updates in Classification and Regression1d ago1d ago
Published inData Science CollectiveEnhancing Time Series Forecasting with Dynamic Weighted TreesA Practical, High-Performance, and Interpretable Decision Tree Approach Incorporating Autoregressive Lags, Seasonal Dynamics, and…5d ago25d ago2
Published inTowards AIAdaptive Decay-Weighted ARMA: A Novel Approach to Time Series ForecastingIntegrating Recency-Based Loss Weighting and Seasonal Feature Tuning for Enhanced Predictive AccuracyApr 12Apr 12
Published inData Science CollectiveAttention-based Neural Network Distillation: Enhancing Performance through Learned Weighted…A Novel Framework Demonstrating Empirical Benefits and Probabilistic Interpretations via Neural Attention in Teacher-Student DistillationApr 10Apr 10
Published inTowards AIBeyond Simple Inversion: Building and Applying Inverse Neural NetworksTheory, training tricks, and real‑world case studies — solving multi‑root equations and beyondApr 73Apr 73
Published inData Science CollectiveTeaching Models to Choose Features Wisely: A Distill-to-Select ApproachA Novel Approach to Enhancing Transparency by Distilling Complex Predictors into Sparse, Human-Readable ModelsApr 21Apr 21
Published inTowards AIReinforcement Learning-Enhanced Gradient Boosting MachinesA Novel Approach to Integrating Reinforcement Learning within Gradient Boosting Internal Optimization for Superior Predictive PerformanceApr 13Apr 13
Published inData Science CollectiveBuilding Smarter Portfolios with Dynamic Programming and Reinforcement Learning and Dynamic…A Novel Framework Integrating Policy Gradient Optimization and State-Driven Buy-and-Hold Strategies for Stock SelectionMar 311Mar 311
Published inTowards AIAdaptive Multi-Teacher Distillation for Enhanced Supervised LearningA Novel Approach for Dynamically Combining Multiple Predictive Models into a Lightweight High-Performance Student ModelMar 244Mar 244
Published inData Science CollectiveStructural Distillation for Cross-Dataset Uplift Modeling with Reinforcement LearningA Novel Approach to Transfer Partial Teacher Model Knowledge from Control to Treatment Data for Rapid AB Testing and Campaign OptimizationMar 241Mar 241
