Research, software, data and methods. Inspect the context, compare your options, and take a useful next step.
UCI Machine Learning Repository
A constructed feature-selection problem for testing signal recovery among distracting dimensions.
Investigate Does feature selection improve Madelon classification when selection is nested inside training folds?
UCI Machine Learning Repository
Simulated telescope event features for studying signal-background discrimination.
Investigate How do classwise calibration and low-false-positive performance compare for two bounded gamma-event baselines?
UCI Machine Learning Repository
Aggregate road-traffic counts and weather context for evaluating demand forecasts over time.
Investigate Does a seasonal traffic baseline outperform a weather-assisted model on a final chronological holdout?
UCI Machine Learning Repository
Metro compressor sensor time series for investigating event detection and forecasting.
Investigate Can a fixed anomaly rule detect compressor events on a held-out chronological segment without label leakage?
UCI Machine Learning Repository
Microturbine operating measurements for evaluating electrical-output prediction across regimes.
Investigate How does electrical-energy prediction generalize to held-out microturbine operating regions?
UCI Machine Learning Repository
Mass-spectrometry descriptors for investigating microbial identification from chemical signals.
Investigate How does MicroMass identification change when related sample measurements are grouped before splitting?
Upstream project repository
Compact software engineering agent implementation for examining minimal agent loops and inspectable action traces.
Investigate Can a tiny offline task distinguish task success from a successful shell command in a recorded agent trajectory?
UCI Machine Learning Repository
Particle-detector event descriptors for studying bounded signal-background classification.
Investigate How does MiniBooNE class calibration vary between a small linear model and a nonlinear baseline?
Upstream project repository
Experiment and model lifecycle tooling for preserving parameters, artifacts and evaluation lineage across runs.
Investigate Can an experiment record reconstruct the exact configuration and artifact digest for a chosen synthetic run?
UCI Machine Learning Repository
Artificial logical concepts for comparing learned rules with documented generating concepts.
Investigate Which MONK concept fails first when irrelevant features are retained under a fixed protocol?
UCI Machine Learning Repository
Mushroom descriptors for auditing categorical prediction and missing-feature dependence.
Investigate Does mushroom class prediction remain accurate after removing the most informative descriptor?
UCI Machine Learning Repository
Molecular conformations and activity labels for studying grouped chemical classification.
Investigate How does molecule-grouped evaluation change musk prediction relative to splitting conformations?
Recorded license evidence is scoped to each source; it is not a blanket permission or a verification result.