Research
Beat a baseline model and your code becomes the latest best.
0 / 5 baselines beaten
Open a challenge to read the notes and run your solution.
Tabular Classification ShowdownTwo noisy 2D blobs, one majority class. Get more labels right than always guessing the majority.accuracy · baseline 0.6667 (Majority class) → beat it20 ptsNonlinear Regression ChaseThe data follows a parabola plus noise. A straight line leaves the curve on the table — add curvature and drive MSE down.MSE · baseline 0.4432 (Linear regression) → beat it25 ptsImbalanced Signal HuntTwenty features, mostly noise, and rare positives. Beat the all-negative baseline by finding the sparse signal.F1 · baseline 0.0000 (Always negative) → beat it30 ptsNoisy Sensor DenoisingEach row is five noisy samples of a triangle-wave sensor ending at time t. Beat the trailing moving average on MSE.MSE · baseline 0.2575 (Moving average) → beat it25 ptsMini Language ModelNext-token prediction over a four-token alphabet generated by a 2nd-order Markov process. Beat the unigram frequency baseline.accuracy · baseline 0.3000 (Unigram frequency) → beat it35 pts