After a Good Fit, What Should We Measure?
A short note on choosing the next observation when several parameter explanations already fit the data.
Topic
Mathematical thinking
Questions, models, geometry and mathematical judgement.
Start with an article
Filter writing on this topic →From Pixels to Patrols: Why Calibration Changes a Synthetic Patrol Plan
A technical reading of an AI4Nature@AVSS 2026 camera-trap study, tracing how probability calibration reaches a synthetic patrol allocator while keeping programme, evidence, and publication status separate.
When Michaelis-Menten Stops Being Valid
A 243-case atlas treats Michaelis-Menten kinetics as a testable model reduction and maps where trajectory or event-time errors become important.
When Publication Count Selects the Method
A cultural-evolution model shows how publication bias and visible output can select low-effort methods, and when replication audits change that selection pressure.
Go a little deeper
Project overviews
2026
Michaelis-Menten QSSA Validity Atlas
A 243-case parameter atlas measures where the Michaelis-Menten quasi-steady-state reduction agrees with full mass-action kinetics and where timing or trajectory error becomes material.
2026
Publish-or-Perish and Method Selection
An agent-based mechanism experiment asks how publication rewards, publication bias, replication audits, and failed-replication penalties select methodological effort.
2026
From Pixels to Patrols
An evidence-bounded editorial record of a calibration-aware synthetic camera-trap pipeline that connects class probabilities to cell-level uncertainty and constrained patrol allocation.