Meta
Meta AI / Reality Labs / Ads ranking. Heavy on PyTorch internals, recommendation systems, multi-task learning.
88 problems tagged for Meta.
Sign in to track your progress0%
Solve a problem (pass its tests) and it's checked off here automatically — your progress is saved, so it stays checked when you come back.
- Adam optimiser stepMedium
- BatchNorm forward (train + eval modes)Medium
- Bayesian A/B Test (Beta-Binomial)Medium
- Binary Quantization + Hamming RetrievalMedium
- BM25 ScoringMedium
- Bootstrap Confidence Interval (Percentile)Medium
- Brier Score + Reliability DecompositionMedium
- Scaled dot-product self-attentionMedium
- Linear forward (Wx + b)Easy
- Chi-Square Test of IndependenceMedium
- CLIP InfoNCE LossMedium
- Clipped Inverse Propensity WeightingMedium
- Continuous Batching SchedulerHard
- Cost-Sensitive Decision ThresholdMedium
- CUPED Variance ReductionMedium
- DDM Concept Drift DetectorMedium
- Delta Method for Ratio Metric VarianceHard
- Difference-in-DifferencesMedium
- Expected Calibration Error (ECE)Medium
- Expected Reciprocal Rank (ERR)Medium
- Feature Hashing (Hashing Trick)Medium
- Fleiss' KappaMedium
- Generalised Advantage Estimation (GAE)Medium
- Gradient Boosting: One Round (Stump)Medium
- Group / Time-Series CV SplitEasy
- Hard Negative Mining for RetrieverMedium
- Hash Bucket A/B AssignmentEasy
- Histogram Split Finding (GBM)Hard
- HNSW Greedy Graph SearchHard
- Holt-Winters Exponential SmoothingMedium
- Post-Training int8 QuantizationMedium
- Team-Draft Interleaving (Online Ranking Eval)Medium
- IPO / SimPO LossMedium
- Isolation Forest Anomaly ScoreMedium
- Isotonic Regression (PAVA)Hard
- IVF (Inverted File) SearchMedium
- Jensen-Shannon Divergence (Drift)Medium
- Knowledge-Distillation LossMedium
- Kolmogorov-Smirnov StatisticMedium
- KTO (Kahneman-Tversky) LossHard
- KV-Cache Eviction (Sliding Window)Medium
- Label-Noise Detection (Confident Learning)Hard
- LambdaMART Lambda GradientsHard
- Levenshtein Edit DistanceEasy
- LinUCB Contextual BanditHard
- ListNet Listwise Ranking LossMedium
- Longest Common SubsequenceMedium
- Magnitude PruningEasy
- Mann-Whitney U TestMedium
- mAP for Object DetectionHard
- Matthews Correlation Coefficient (MCC)Easy
- Mean Average Precision (MAP)Medium
- MinHash for Near-Duplicate DetectionMedium
- MIPS -> L2 ReductionMedium
- Multi-Task Score FusionMedium
- Multiple-Testing Correction (Bonferroni / BH-FDR)Medium
- Negative Sampling (Implicit Feedback)Medium
- Non-Max Suppression (NMS)Medium
- Out-of-Fold Target EncodingMedium
- ORPO Loss (Odds-Ratio Preference Optimisation)Medium
- Paged-Attention KV Block AllocationMedium
- Pairwise Ranking Loss (BPR / RankNet)Medium
- Permutation Feature ImportanceMedium
- PII Detection & RedactionMedium
- Pinball (Quantile) Regression LossMedium
- Platt ScalingMedium
- Point-in-Time Feature JoinMedium
- Population Stability Index (PSI)Medium
- PR-AUC (Average Precision)Medium
- Product Quantization (PQ) Encode + ADCHard
- Random-Hyperplane LSH (Cosine)Medium
- Reward-Model Margin LossMedium
- Sample Ratio Mismatch (SRM) CheckMedium
- Sample Size for Statistical PowerMedium
- In-Batch Sampled Softmax + logQ CorrectionHard
- Sequential Test (mSPRT-Style Always-Valid p-Value)Hard
- Shadow / Canary Rollback DecisionMedium
- Sliding-Window Counter (Streaming Feature)Medium
- Smoothed Target (Mean) EncodingMedium
- SMOTE OversamplingMedium
- Softmax from scratchEasy
- Temporal Train/Val SplitEasy
- Thompson Sampling (Bernoulli-Beta)Medium
- Two-Sample Test + p-valueMedium
- Two-Tower Top-k RetrievalEasy
- ViT Patch EmbeddingMedium
- Wilson Confidence Interval (Proportion)Easy
- WordPiece Tokenization (Greedy Longest-Match)Medium