In Physics
A property of a dynamical system where time averages equal space averages over invariant sets.
In AI & ML
The property where every state can be reached from every other state with non-zero probability.
Know first
Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.
Dynamical system2Fourier series1Hamiltonian mechanics1Markov chain1Measure theory1Phase space1Probability1Probability theory1
Opens up
Concepts that name this one as a prerequisite.
In Physics
3Each in their own terms, from the easiest level.
Alexander Polyakov
The system occupies with equal probability all allowed places.Qian Hong
An assumption that data collected forever exhibits stabilized statistical patterns, framed here as an assumption on data rather than the system itself.
In AI & ML
1Lectures that assume it
1Connections
11The lectures that define it or assume it, and the concepts next to it.
Concepts5
Lectures6
- Policy Gradient Foundations: Trajectory Distributions & High-Dimensional State Spaces
- Fluctuation-Dissipation Relations, Langevin Dynamics & Fokker-Planck Equations
- Dynamical Asymmetry is C^1-Typical
- Hamiltonian Mechanics, Symplectic Structure & Canonical Transformations
- Statistical Thermodynamics as a Paradigm for Data Science
- Many-Body Localization and its Discontents