Skip to content
Home / Catalog / Stable Latent-State Fitting for Long Symbolic Sequences

Stable Latent-State Fitting for Long Symbolic Sequences

  • Task ID: computer_science.stable_latent_state_fitting
  • Domain: computer_science
  • Subdomain: machine_learning_sequence_models
  • Status: final
  • Benchmark set: seed42 (60 tasks)
  • Tags: latent_state_models, sequence_modeling, numerical_stability, categorical_data, expectation_maximization, posterior_diagnostics

Runtime and requirements

  • Estimated time: 60-120 minutes
  • Python: >=3.10
  • Packages: numpy>=1.24, pandas>=2.0, scipy>=1.11, matplotlib>=3.7
  • GPU required: no
  • Network required: no

Public input and output contract

Inputs

  • metadata.json (data): Agent-facing schema, alphabet labels, row-order convention, and sequence summaries.
  • sequences.csv (data): Long categorical observation streams with columns global_index, sequence_id, position, symbol, and symbol_label.

Outputs

  • analysis.py (code): Agent analysis script that reads data/ and writes all results.
  • results/candidate_summary.csv (data): Candidate-size comparison and selected-row summary.
  • results/matrix_a.csv (data): First normalized square table in long format.
  • results/matrix_b.csv (data): Second normalized row-by-symbol table in long format.
  • results/vector_a.csv (data): Normalized starting-row table.
  • results/fit_trace.csv (data): Iterative fitting objective trace.
  • results/local_weights.npy (data): Position-wise normalized weights in data/sequences.csv global_index order.
  • results/neighbor_summary.csv (data): Adjacent-row summary table.
  • results/local_assignments.csv (data): One integer row label for each sequence position.
  • results/stream_diagnostics.csv (data): Per-sequence objective, uncertainty, and occupancy diagnostics.
  • results/diagnostics.png (figure): Diagnostic figure summarizing fit trace, row profiles, occupancies, and assignments.

Public repository files

View this task in ASI-Bench

Formal benchmark task directories in the public repository are metadata-only. Versioned prompts and inputs are distributed through the pinned benchmark dataset.

Provenance

This task is included in the current seed42 benchmark set at dataset revision f11a199f71fb4b854ac43a1bf548d5df519141a9.

This page is generated only from files tracked in the public ASI-Bench repository at commit f18382f03faf. Submission bundles, run logs, private scoring configuration, and private reference answers are not read by this page generator.