0/ 30 SOLVED
CLASSICAL ML0/10
DEEP LEARNING0/10
TRANSFORMERS0/8
INTERPRETABILITY0/2
0 wonML-ARENA BENCHMARKS
StatusPROJECTTOPICS (YELLOW = NEW)BENCHMARKINDUSTRYMINEEST.
BL-01.Bike-demand predictor
MLPYTHONafter original
NEW ml.linregNEW ml.featuresNEW ml.evaltest RMSEsklearn LinearRegression—5 h
BL-02.SMS spam filter
MLPYTHONafter original
NEW ml.naive-bayesml.evalml.featuresF1 on held-out messagessklearn MultinomialNB—5 h
BL-03.Rain-tomorrow forecaster
MLPYTHONafter original
NEW ml.logregml.evalml.featureslog-loss; calibrationsklearn LogisticRegression—5 h
BL-04.Recipe recommender
MLPYTHONafter originalFITS NOW
NEW ml.knnml.featuresml.evalrecall@10; queries per secondsklearn NearestNeighbors—5 h
BL-05.Loan decision explainer
MLPYTHONafter Breiman et al.
NEW ml.treesml.evalml.featuresaccuracy at the same depthsklearn DecisionTreeClassifier—7 h
BL-06.Colour-palette extractor
MLPYTHONafter original
NEW ml.clusteringml.knnml.featuresinertiasklearn KMeans (n_init=10)—5 h
BL-07.Map of anything
MLPYTHONafter van der Maaten & HintonFITS NOW
NEW ml.dimredml.clusteringml.evalPCA output; neighbourhood trustworthinesssklearn PCA · sklearn TSNE—6 h
BL-08.Face-or-not detector
MLPYTHONafter original
NEW ml.svm-kernelsml.evalml.dimredaccuracysklearn SVC—7 h
BL-09.Tiny gradient booster
MLPYTHONafter original
ml.treesml.evalml.featurestest accuracysklearn HistGradientBoostingClassifier—9 h
BL-10.AutoML-lite
MLPYTHONafter original
ml.evalml.featuresml.treesml.naive-bayesml.svm-kernelsCV scores; leakage testssklearn cross_val_score—8 h
BL-11.Neural net from scratch
MLPYTHONafter Andrej Karpathy
NEW dl.mlpNEW dl.backpropml.logreggradients; accuracyfinite differences—6 h
BL-12.Deep-net rescue
MLPYTHONafter original
NEW dl.init-normNEW dl.optimdl.backproptraining curvePyTorch, same setup—7 h
BL-13.Flower classifier
MLPYTHONafter original
NEW dl.regularisedl.mlpml.evalvalidation accuracyPyTorch reference model—6 h
BL-14.Convolution engine
MLPYTHONafter original
NEW dl.cnnNEW sys.kernelsdl.backpropconv outputs; speeduptorch.nn.functional.conv2d—9 h
BL-15.Style writer
MLPYTHONafter Andrej Karpathy
NEW dl.rnnNEW tf.lmdl.backpropgradients; validation lossPyTorch LSTM · n-gram baseline—8 h
BL-16.Word map
MLPYTHONafter Mikolov et al.
NEW dl.embeddingsNEW tf.tokenizeml.knnanalogy accuracygensim Word2Vec, same corpus—8 h
BL-17.Spoken-digit recogniser
MLPYTHONafter original
NEW dl.audio-tsdl.cnndl.rnnml.evaltest accuracylibrosa spectrograms · measured—7 h
BL-18.Sprite generator
MLPYTHONafter original
NEW dl.generativedl.cnndl.optimreconstruction loss; sample gridPyTorch reference VAE—10 h
BL-19.Game-playing agent
MLPYTHONafter Mnih et al.
NEW dl.rldl.mlpdl.optimaverage return on CartPole-v1Gymnasium solved threshold: 475—8 h
BL-20.Paper recommender
MLPYTHONafter Kipf & Welling
NEW dl.gnndl.embeddingsml.evalnode-classification accuracy on CoraGCN paper: 81.5%—8 h
BL-21.Tiny GPT
MLPYTHONafter Andrej Karpathy
NEW tf.attentiontf.lmtf.tokenizedl.init-normdl.regularisevalidation lossnanoGPT, same data and size—12 h
BL-22.Fast GPT
MLPYTHONafter original
NEW tf.inferenceNEW sys.memorytf.attentionspeed; output match; perplexityyour uncached fp32 model—8 h
BL-23.Faster training
MLPYTHONafter original
NEW tf.trainingNEW sys.paralleldl.optimtime to target lossyour fp32 baseline · measured—8 h
BL-24.Teach GPT-2 a trick
MLPYTHONafter Hu et al.
NEW tf.finetuneNEW tf.evaltf.lmtf.trainingeval loss; trainable paramsfull fine-tune · measured—8 h
BL-25.Ask your notes
MLPYTHONafter original
NEW tf.retrievaltf.evaldl.embeddingsml.knnrecall@5 on 50 questions you writeBM25 (rank_bm25)—8 h
BL-26.Search photos by text
MLPYTHONafter original
NEW tf.multimodaldl.embeddingstf.retrievalrecall@5pretrained CLIP zero-shot · measured—9 h
BL-27.Serve your GPT
MLPYTHONafter original
NEW sys.servingtf.inferencesys.parallelsys.memoryp50/p95 latency; throughputone-request-at-a-time server—7 h
BL-28.Mini compiler
MLPYTHONafter original
NEW sys.compilerssys.kernelsdl.backpropspeed; output matchyour unfused version—10 h
BL-29.X-ray your GPT
MLPYTHONafter nostalgebraist
NEW interp.lensNEW interp.probesNEW interp.circuitstf.attentionml.dimredprobe accuracy; induction scoresTransformerLens on GPT-2 small—10 h
BL-30.Feature finder and steering wheel
MLPYTHONafter Anthropic
NEW interp.saeNEW interp.steeringtf.lmml.clusteringreconstruction error and sparsity; steering successSAELens defaults, same activations—12 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.