🧠 Loading live leaderboard...
BTC paper portfolios. Counts load with the snapshot. Paper portfolios are separate from shadow models. Paper participation does not establish model readiness.
≠ AI Arena (LLM decision-makers · here: statistical models)
Statistical models in BTC paper simulation
Inference, training and historical score are three separate things. A value missing from the snapshot is shown as N/A.
Schematic of the principle only. The observed state of inference and training is in the status block above.
BTC paper simulation. Loading dated snapshot...
ML Arena presents BTC paper portfolios and separate shadow models; their number is the one in the snapshot shown above. An artifact's presence establishes neither readiness nor recent inference. Unverified metrics remain unknown.
See the complète AI vs ML benchmark and the free training course.
Affiliate links — we may earn a small commission
The number of models, the time of the last inference and the eligibility of the next training run come from the snapshot loaded by this page. Paper P&L and win rate are historical scores of a simulation.
This page states no bar interval and no horizon in hours for the models: the snapshot gives declared parameters, and the timeframe of past training is shown as reported, including when it is unknown.
XGBoost gagne encore 80% des competitions Kaggle quant. Le deep learning est sexy mais perd souvent face a un GradientBoosting bien tune sur features.