<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Trajectory on 3rd layer</title><link>https://3rdlayer.uk/tags/trajectory/</link><description>Recent content in Trajectory on 3rd layer</description><generator>Hugo -- 0.157.0</generator><language>en-US</language><lastBuildDate>Sun, 27 Sep 2026 09:00:00 +0900</lastBuildDate><atom:link href="https://3rdlayer.uk/tags/trajectory/index.xml" rel="self" type="application/rss+xml"/><item><title>Trajectory ML 09: Learn where a state moves next</title><link>https://3rdlayer.uk/posts/trajectory-ml-09-velocity/</link><pubDate>Sun, 27 Sep 2026 09:00:00 +0900</pubDate><guid>https://3rdlayer.uk/posts/trajectory-ml-09-velocity/</guid><description>Estimate a velocity field from paired observations, then discover why predicting one step and rolling out a trajectory are different tests.</description></item><item><title>Trajectory ML 10: Move a population with optimal transport</title><link>https://3rdlayer.uk/posts/trajectory-ml-10-transport/</link><pubDate>Sun, 27 Sep 2026 09:00:00 +0900</pubDate><guid>https://3rdlayer.uk/posts/trajectory-ml-10-transport/</guid><description>Infer a mass-preserving coupling between snapshots, distinguish it from tracked identity, and implement Sinkhorn transport.</description></item><item><title>Trajectory ML 11: Continuous dynamics and flow matching</title><link>https://3rdlayer.uk/posts/trajectory-ml-11-continuous/</link><pubDate>Sun, 27 Sep 2026 09:00:00 +0900</pubDate><guid>https://3rdlayer.uk/posts/trajectory-ml-11-continuous/</guid><description>Connect neural vector fields, ODEs, SDEs, and flow matching without confusing generated paths with observed temporal mechanisms.</description></item><item><title>Trajectory ML 12: Outcomes, survival, and causal questions</title><link>https://3rdlayer.uk/posts/trajectory-ml-12-fate-causality/</link><pubDate>Sun, 27 Sep 2026 09:00:00 +0900</pubDate><guid>https://3rdlayer.uk/posts/trajectory-ml-12-fate-causality/</guid><description>Predict outcomes without future leakage, handle censored follow-up, and distinguish forecasting from the effect of an intervention.</description></item></channel></rss>