MétaCan
Menu
Back to cohort
Record W3085735333 · doi:10.48550/arxiv.2009.06799

The Importance of Pessimism in Fixed-Dataset Policy Optimization

2020· preprint· en· W3085735333 on OpenAlexaff
Jacob Buckman, Carles Gelada, Marc G. Bellemare

Bibliographic record

VenuearXiv (Cornell University) · 2020
Typepreprint
Languageen
FieldDecision Sciences
TopicAdvanced Bandit Algorithms Research
Canadian institutionsMcGill UniversityUniversité de Montréal
Fundersnot available
KeywordsPessimismCore (optical fiber)Computer scienceValue (mathematics)Order (exchange)Mathematical optimizationArtificial intelligenceEconometricsEconomicsMachine learningMathematics

Abstract

fetched live from OpenAlex

We study worst-case guarantees on the expected return of fixed-dataset policy optimization algorithms. Our core contribution is a unified conceptual and mathematical framework for the study of algorithms in this regime. This analysis reveals that for naive approaches, the possibility of erroneous value overestimation leads to a difficult-to-satisfy requirement: in order to guarantee that we select a policy which is near-optimal, we may need the dataset to be informative of the value of every policy. To avoid this, algorithms can follow the pessimism principle, which states that we should choose the policy which acts optimally in the worst possible world. We show why pessimistic algorithms can achieve good performance even when the dataset is not informative of every policy, and derive families of algorithms which follow this principle. These theoretical findings are validated by experiments on a tabular gridworld, and deep learning experiments on four MinAtar environments.

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.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.030
metaresearch head score (Gemma)0.177
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Theoretical or conceptual · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.030
Threshold uncertainty score0.159

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0300.177
Meta-epidemiology (narrow)0.0020.001
Meta-epidemiology (broad)0.0030.002
Bibliometrics0.0020.002
Science and technology studies0.0020.006
Scholarly communication0.0060.010
Open science0.0030.004
Research integrity0.0030.008
Insufficient payload (model declined to judge)0.0030.001

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 head0.244
GPT teacher head0.338
Teacher spread0.094 · 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

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designTheoretical or conceptual
Domainnot available
GenreMethods

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations23
Published2020
Admission routes1
Has abstractyes

Explore more

Same venuearXiv (Cornell University)Same topicAdvanced Bandit Algorithms ResearchFrench-language works237,207