Task 008: Pre-Experiment Plan¶
Priority: MEDIUM Status: IN PROGRESS Source: Weekly catchup 2026-03-22, Telegram discussion
Overview¶
Strategic tasks that set up the next phase of research. Not urgent for tomorrow's meeting but important for the week ahead.
Checklist¶
DMC vs ARD Comparison (Issue #6)¶
- Run all 33 experiments with DMC tracking enabled
- Generate side-by-side ranking: DMC vs ARD for each method
- Identify cases where rankings differ (these are the interesting ones)
- Write up comparison as a finding
- Comment on Issue #6 with results
Sparse Parity as RL Environment¶
- Design env interface: observation space, action space, reward signal
- Observation: problem spec (n, k, metric type) + current best score
- Action: algorithm choice + hyperparameter selection
- Reward: improvement in DMC/ARD over baseline
- Prototype
src/sparse_parity/rl_env.pywith Gymnasium interface - Test: can a simple agent (random search) find GF(2) through the env?
- Document as a finding if the 33 experiments serve as ground truth
Public Domain License¶
- Add LICENSE file (CC0-1.0 or Unlicense) to repo root
- Confirm with Yaroslav on preferred license text
- Update README if needed
Lukas Kaiser / Mar 30 Meeting Prep¶
- Ensure docs site is up to date before Mar 30
- Review survey page for completeness
- Prepare a 5-minute "state of research" summary for newcomers
Context¶
Yaroslav's idea (chat-yaroslav, Mar 21): wrap challenges into RL environments for Anthropic/PrimeIntellect. Our 33 experiments serve as an answer key -- richer signal than most RL envs.
PrimeIntellect research grants: compute + stipends for novel environments.