persistence-agent 0.1.0

Persistent homology for agent behavior profiling — topological fingerprints
Documentation
  • Coverage
  • 45.83%
    33 out of 72 items documented1 out of 33 items with examples
  • Size
  • Source code size: 40.59 kB This is the summed size of all the files inside the crates.io package for this release.
  • Documentation size: 934.47 kB This is the summed size of all files generated by rustdoc for all configured targets
  • Ø build duration
  • this release: 8s Average build duration of successful builds.
  • all releases: 8s Average build duration of successful builds in releases after 2024-10-23.
  • Links
  • SuperInstance/persistence-agent
    0 0 0
  • crates.io
  • Dependencies
  • Versions
  • Owners
  • SuperInstance

persistence-agent

Persistent homology for agent behavior profiling — topological fingerprints.

This crate computes persistent homology from point clouds representing agent state trajectories and maps topological invariants to behavior archetypes.

Quick Start

use persistence_agent::agent_features::AgentProfiler;

// Agent observations over time (e.g. 2D state vectors)
let obs = vec![
    vec![0.0, 0.0],
    vec![0.1, 0.0],
    vec![0.0, 0.1],
    vec![1.0, 1.0],
    vec![1.1, 1.0],
];

let profiler = AgentProfiler::new(1);
let profile = profiler.profile(obs).unwrap();
println!("Archetype: {}", profile.archetype);
println!("Entropy:   {:.3}", profile.persistence_entropy);

Pipeline

  1. PointCloud — embed observations, compute distances
  2. VietorisRipsComplex — build filtration from pairwise distances
  3. BoundaryMatrix — mod-2 boundary matrix + column reduction
  4. Barcode — extract persistence pairs and Betti curves
  5. AgentProfiler — classify behavior into archetypes

Archetypes

Archetype Topological Signature
Steady Single persistent cluster (β₀ = 1)
Explorer Many short-lived loops
Volatile Many disconnected components
Deep Long-lived higher-dimensional features
Balanced Mixed features, no dominant signature