MétaCan
Menu
Back to cohort
Record W4386126338 · doi:10.1007/s10994-023-06368-z

Cautious policy programming: exploiting KL regularization for monotonic policy improvement in reinforcement learning

2023· article· en· W4386126338 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.

Bibliographic record

VenueMachine Learning · 2023
Typearticle
Languageen
FieldComputer Science
TopicReinforcement Learning in Robotics
Canadian institutionsUniversity of Alberta
FundersJapan Society for the Promotion of Science
KeywordsReinforcement learningMonotonic functionComputer scienceMathematical optimizationRegularization (linguistics)Bellman equationEntropy (arrow of time)MaximizationUpper and lower boundsExploitArtificial intelligenceMathematics

Abstract

fetched live from OpenAlex

Abstract In this paper, we propose cautious policy programming (CPP), a novel value-based reinforcement learning (RL) algorithm that exploits the idea of monotonic policy improvement during learning. Based on the nature of entropy-regularized RL, we derive a new entropy-regularization-aware lower bound of policy improvement that depends on the expected policy advantage function but not on state-action-space-wise maximization as in prior work. CPP leverages this lower bound as a criterion for adjusting the degree of a policy update for alleviating policy oscillation. Different from similar algorithms that are mostly theory-oriented, we also propose a novel interpolation scheme that makes CPP better scale in high dimensional control problems. We demonstrate that the proposed algorithm can trade off performance and stability in both didactic classic control problems and challenging high-dimensional Atari games.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.002
metaresearch head score (Gemma)0.002
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesMeta-epidemiology (narrow)
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Methods · Consensus signal: none
Teacher disagreement score0.969
Threshold uncertainty score1.000

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0020.002
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0020.003
Science and technology studies0.0010.000
Scholarly communication0.0000.001
Open science0.0010.001
Research integrity0.0000.001
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.018
GPT teacher head0.285
Teacher spread0.267 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it