Dataset Viewer
Auto-converted to Parquet Duplicate
Search is not available for this dataset
learnability
dict
continual_pressure
dict
bounded_replay_value
dict
replay_scaling
dict
residual_headroom
dict
{ "joint_every_seed_min": 80, "joint_cohort_mean_min": 82, "pass_every_seed": false, "pass_cohort_mean": false }
{ "joint_minus_sgd_every_seed_min": 55, "joint_minus_sgd_cohort_mean_min": 60, "pass_every_seed": true, "pass_cohort_mean": true }
{ "er200_minus_sgd_every_seed_min": 15, "er200_minus_sgd_cohort_mean_min": 20, "pass_every_seed": true, "pass_cohort_mean": true }
{ "er500_minus_er200_every_seed_min": 5, "er500_minus_er200_cohort_mean_min": 8, "er1000_minus_er500_every_seed_min": 5, "er1000_minus_er500_cohort_mean_min": 8, "pass_er500_every_seed": true, "pass_er500_mean": true, "pass_er1000_every_seed": true, "pass_er1000_mean": true }
{ "joint_minus_er1000_every_seed_min": 4, "joint_minus_er1000_cohort_mean_min": 7, "pass_every_seed": false, "pass_cohort_mean": false }

MNIST-360 Continual-Learning Qualification

A reproducible qualification study of the actual MNIST-360 General Continual Learning stream using a pinned official Mammoth implementation.

The study prospectively compared five fixed conditions across 12 fresh seeds:

Condition Mean final accuracy
SGD 20.37%
ER-200 47.91%
ER-500 61.18%
ER-1000 75.17%
Joint/offline 81.67%

Result

Negative qualification: NOT_SUFFICIENTLY_LEARNABLE_UNDER_FROZEN_GATES

The first binding gate required joint/offline training to reach at least 80% on every seed and 82% cohort mean. The observed joint mean was 81.67%, with two seeds below 80%.

Other pressures were clearly present: ordinary online SGD was poor, bounded replay added substantial value, and larger replay buffers improved performance. But the substrate failed the prospectively frozen learnability criterion, so the negative result was preserved rather than repaired after inspection.

Reproduce

Open notebooks/RUN_IN_COLAB.ipynb in Google Colab. The notebook clones Mammoth at commit e75a491c69fd729edeb01431afb753d9157d9a81, verifies pinned Git objects, applies and hashes the frozen compatibility patch, runs 60 scientific runs, and evaluates the already-frozen gates.

The public notebook changes only release naming and output paths from the source research artifact. Scientific settings, seeds, thresholds, commits, and run configurations are unchanged.

Repository contents

  • protocol/FROZEN_PROTOCOL_PUBLIC_RELEASE.md — prospective frozen protocol, public naming
  • notebooks/RUN_IN_COLAB.ipynb — reproducible Colab runner
  • results/raw_results.csv — all 60 observed final accuracies
  • results/metrics_by_seed.csv — per-seed derived metrics
  • results/cohort_summary.csv — cohort descriptive statistics
  • results/gate_evaluation.json — frozen gate evaluation
  • results/qualification_result.md — concise result and claim boundary
  • docs/SOURCE_CUSTODY.md — source hashes and release modifications

Scope

This artifact evaluates one pinned MNIST-360 substrate and configuration matrix. It does not establish general superiority or inferiority of replay, joint training, SGD, or continual-learning methods.

Downloads last month
41