# Measuring the Right Thing Measurement is needed to *describe* our world *quantitatively*. Measurement gets us out of the qualitative (that building is large) to quantitative (that building is 50 meters tall) empirically The art of doing science and engineering ### General Thoughts 1. [Good Science is about Measurement](Good%20Science%20is%20about%20Measurement.md) 2. [Doing Good Science](Doing%20Good%20Science.md) 3. [Asking Better Questions](Asking%20Better%20Questions.md) 4. [Scientific Measurements Use Chains of Proxies](Scientific%20Measurements%20Use%20Chains%20of%20Proxies.md) 5. [Explanationless Prediction Is Impossible](Explanationless%20Prediction%20Is%20Impossible.md) 6. [Looking Under the Lamppost Because it is Brighter There](Looking%20Under%20the%20Lamppost%20Because%20it%20is%20Brighter%20There.md) ### Favorite Measurements 1. [Linear Mode Connectivity and the Lottery Ticket Hypothesis](Linear%20Mode%20Connectivity%20and%20the%20Lottery%20Ticket%20Hypothesis.pdf) 2. [Out of Model Scope](Out%20of%20Model%20Scope.md) 3. Revising the Evolutionary Computation Abstraction: Minimal Criteria Novelty Search 4. [Understanding Transformers via N-gram Statistics](Understanding%20Transformers%20via%20N-gram%20Statistics.pdf) ### Personal Measurements 1. [Explainability - Anomalies and Model Scope](Explainability%20-%20Anomalies%20and%20Model%20Scope.md) 2. [Backtest Uncertainty](Backtest%20Uncertainty.md) 3. [Continuous Rank Probability Score](Continuous%20Rank%20Probability%20Score.md) 4. [Critical Classification Scores](Critical%20Classification%20Scores.md) 5. Cover 1. [Cover](Cover.md) 2. [Cover (Active)](Cover%20(Active).md) 3. [Explanation of how Cover Works](Explanation%20of%20how%20Cover%20Works.md) (hedging clustering analysis ) TODO: * KDE score approach tom used [here](https://zero-g-capital.slack.com/archives/C04ANJFNFE2/p1711129166107109) * Overfitting sweep * risk via constraint space --- Date: 20240326 Links to: Tags: References: * []()