MétaCan
Menu
Back to cohort
Record W4382239271 · doi:10.1609/aaai.v37i6.25864

The Sufficiency of Off-Policyness and Soft Clipping: PPO Is Still Insufficient according to an Off-Policy Measure

2023· article· en· W4382239271 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.
fundA Canadian funder is recorded on the work.

Bibliographic record

VenueProceedings of the AAAI Conference on Artificial Intelligence · 2023
Typearticle
Languageen
FieldComputer Science
TopicOptimization and Search Problems
Canadian institutionsUniversity of Alberta
FundersNatural Sciences and Engineering Research Council of CanadaNational Natural Science Foundation of ChinaAlberta Machine Intelligence Institute
KeywordsClipping (morphology)Computer scienceSpace (punctuation)Metric (unit)Measure (data warehouse)Mathematical optimizationSigmoid functionCode (set theory)Function (biology)AlgorithmMathematicsData miningArtificial intelligenceEconomicsOperations managementProgramming language

Abstract

fetched live from OpenAlex

The popular Proximal Policy Optimization (PPO) algorithm approximates the solution in a clipped policy space. Does there exist better policies outside of this space? By using a novel surrogate objective that employs the sigmoid function (which provides an interesting way of exploration), we found that the answer is "YES", and the better policies are in fact located very far from the clipped space. We show that PPO is insufficient in "off-policyness", according to an off-policy metric called DEON. Our algorithm explores in a much larger policy space than PPO, and it maximizes the Conservative Policy Iteration (CPI) objective better than PPO during training. To the best of our knowledge, all current PPO methods have the clipping operation and optimize in the clipped policy space. Our method is the first of this kind, which advances the understanding of CPI optimization and policy gradient methods. Code is available at https://github.com/raincchio/P3O.

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 categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.809
Threshold uncertainty score0.700

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0020.002
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.003
Science and technology studies0.0010.000
Scholarly communication0.0010.000
Open science0.0030.001
Research integrity0.0000.000
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 head0.105
GPT teacher head0.343
Teacher spread0.237 · 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