Every Curve Is a Compressed Argument: Key Data Science Plots Explained
burny_tech · x · 2026-11-12
A widely shared thread explaining what common ML/statistics plots actually encode: the KS statistic is the maximum vertical gap between two cumulative curves; SHAP assigns each feature an additive contribution to a prediction; ROC traces true-positive vs false-positive rate as the threshold sweeps 0 to 1; elbow and silhouette mark the cluster count where inertia flattens or separation peaks; bias squared falls while variance climbs with complexity, with total error minimized where they balance. The kicker: every curve is a compressed argument.
More from Research
- Qwen 3.8 27B on a laptop can read model weights, replicating the Weight-Space Programming paper — dl_weekly · 2026-11-12
- Trask: Transformer weight columns are concept detectors, but many concepts remain missing — iamtrask · 2026-11-12
- Dense rewards lose to standard GRPO at matched compute, arXiv paper finds — burny_tech · 2026-11-12
- Wayve unveils Learned Safety Envelope: end-to-end trained live risk prediction for AI driving — alexgkendall · 2026-11-12
- HipKittens turns one: breaking AMD's 'CUDA moat' for a multi-silicon future — simran_s_arora · 2026-11-12
- Why world model teams pay for gameplay with inputs attached: actions are the scarce half — KnowledgeOk7634 · 2026-11-12