0/ 30 SOLVED
LINEAR ALGEBRA0/12
CALCULUS0/10
PROBABILITY0/8
0 wonML-ARENA BENCHMARKS
StatusPROJECTTOPICS (YELLOW = NEW)BENCHMARKINDUSTRYMINEEST.
BM-01.Note finder
MATHPYTHONafter originalFITS NOW
NEW la.vectorsNEW la.tensorstop-5 matches for 20 queriessklearn cosine_similarity—4 h
BM-02.Matmul speed ladder in Python
MATHPYTHONafter Simon BoehmFITS NOW
NEW la.matmulla.tensorsla.vectorsspeedup over your naive loop, 512 × 512 float32NumPy ~7,900× (measured on one machine; mlarena bench measures yours)—6 h
BM-03.Photo transformer
MATHPYTHONafter originalFITS NOW
NEW la.det-inversela.matmulla.vectorsoutput image vs referencescipy.ndimage.affine_transform—6 h
BM-04.Chemical equation balancer
MATHPYTHONafter originalFITS NOW
NEW la.systemsla.det-inversela.matmul50 textbook equationsSymPy nullspace—5 h
BM-05.Trend fitter
MATHPYTHONafter originalFITS NOW
NEW la.leastsqla.systemsla.matmulfitted coefficientsnumpy.linalg.lstsq—5 h
BM-06.Fit-explosion detector
MATHPYTHONafter originalFITS NOW
NEW la.conditioningla.leastsqla.systemscondition numbers and fit errornumpy.linalg.cond and numpy.polyfit—5 h
BM-07.Long-run predictor
MATHPYTHONafter originalFITS NOW
NEW la.eigenla.matmulla.vectorsdominant eigenvalue and eigenvectornumpy.linalg.eig—5 h
BM-08.Photo shrinker
MATHPYTHONafter originalFITS NOW
NEW la.svdla.eigenla.matmulsingular values; quality at 10% sizenumpy.linalg.svd · JPEG measured locally—6 h
BM-09.Lookalike finder
MATHPYTHONafter Turk & PentlandFITS NOW
la.svdla.eigenla.vectorsla.tensorsreconstruction at k = 20sklearn.decomposition.PCA—6 h
BM-10.Wikipedia ranker
MATHPYTHONafter Nayuki
NEW la.sparse-graphsNEW prob.basicsla.eigentop-100 rankingnetworkx.pagerank—8 h
BM-11.Heat-spread simulator
MATHPYTHONafter originalFITS NOW
NEW la.iterativela.sparse-graphsla.systemssolution and time on a 500×500 gridscipy.sparse.linalg.spsolve—7 h
BM-12.Mini 3D renderer
MATHPYTHONafter originalFITS NOW
la.matmulla.vectorsla.det-inversela.tensorsprojected verticesscipy.spatial.transform.Rotation—8 h
BM-13.Speedometer from GPS
MATHPYTHONafter original
NEW calc.derivativesla.vectorsla.leastsqspeed error on a synthetic tracknumpy.gradient—5 h
BM-14.Terrain explorer
MATHPYTHONafter original
NEW calc.gradientscalc.derivativesla.tensorsgradient field; peak listnumpy.gradient · scipy.ndimage.maximum_filter—5 h
BM-15.Optimizer race
MATHPYTHONafter original
NEW calc.gdcalc.gradientsla.vectorssteps to converge on Rosenbrockscipy.optimize.minimize (BFGS) · measured—6 h
BM-16.Robot arm reacher
MATHPYTHONafter original
NEW calc.chain-jacobiancalc.gdla.systemsreach error on 1,000 random targetsclosed-form inverse kinematics—7 h
BM-17.Derivative calculator
MATHPYTHONafter Max Slater
NEW calc.autodiffcalc.chain-jacobiancalc.derivativesderivatives on 100 formulasSymPy diff · finite differences—7 h
BM-18.Loan and investment solver
MATHPYTHONafter original
NEW calc.hessiancalc.gdla.systemsrate and iterationsscipy.optimize.newton—6 h
BM-19.Budget planner
MATHPYTHONafter original
NEW calc.convexla.systemscalc.gradientsoptimal costscipy.optimize.linprog—7 h
BM-20.Solar-system simulator
MATHPYTHONafter original
NEW calc.odesNEW calc.integrationla.vectorsenergy drift over 100 yearsscipy.integrate.solve_ivp—8 h
BM-21.Guitar tuner
MATHPYTHONafter Jez Swanson
NEW calc.fourierla.matmulla.tensorsFFT output; detected pitchnumpy.fft · labelled clips—7 h
BM-22.Gravity from a video
MATHPYTHONafter original
calc.odescalc.gdcalc.autodiffla.leastsqestimated g9.81 m/s²—10 h
BM-23.Board-game odds engine
MATHPYTHONafter original
NEW prob.samplingprob.basicsla.eigensquare-landing probabilitiesexact Markov-chain answer—6 h
BM-24.Fake-data generator
MATHPYTHONafter original
NEW prob.distributionsNEW prob.expectationprob.samplingcalc.integrationgoodness of fitscipy.stats (KS test)—6 h
BM-25.A/B test calculator
MATHPYTHONafter original
NEW prob.testingNEW prob.bayesprob.distributionsprob.expectationp-values, intervals, posteriorsscipy.stats · conjugate formulas—6 h
BM-26.Text compressor
MATHPYTHONafter original
NEW prob.infoprob.basicsprob.distributionscompressed sizezlib · measured locally—8 h
BM-27.Best-fit distribution finder
MATHPYTHONafter original
NEW prob.mleprob.distributionscalc.gdcalc.hessiancalc.convexprob.infofitted parameters and standard errorsscipy.stats .fit—6 h
BM-28.Phone tracker
MATHPYTHONafter Tim Babb
NEW prob.state-spacela.matmulla.det-inverseprob.distributionsposition errorraw GPS · filterpy—8 h
BM-29.How many taxis?
MATHPYTHONafter original
NEW prob.mcmcprob.bayesprob.samplingposteriorexact grid posterior—7 h
BM-30.Smart autotuner
MATHPYTHONafter original
NEW prob.gpla.systemscalc.gdprob.bayesGP prediction; experiments to find the bestsklearn GaussianProcessRegressor · grid search—10 h

INDUSTRY is the library, paper or record you measure against. MINE is your latest recorded result. Base Form projects add one or two new topics each and reuse the rest.