MétaCan
Menu
Back to cohort
Record W3169514089

EMaQ: Expected-Max Q-Learning Operator for Simple Yet Effective Offline and Online RL

2021· article· en· W3169514089 on OpenAlexaff
Seyed Kamyar Seyed Ghasemipour, Dale Schuurmans, Shixiang Gu

Bibliographic record

VenueInternational Conference on Machine Learning · 2021
Typearticle
Languageen
FieldComputer Science
TopicReinforcement Learning in Robotics
Canadian institutionsGoogle (Canada)University of AlbertaUniversity of Toronto
Fundersnot available
KeywordsReinforcement learningComputer scienceBackupHeuristicOperator (biology)Online and offlineOffline learningSimple (philosophy)Artificial intelligenceGenerative grammarMachine learningMathematical optimizationTheoretical computer scienceOnline learningMathematics
DOInot available

Abstract

fetched live from OpenAlex

Off-policy reinforcement learning (RL) holds the promise of sample-efficient learning of decision-making policies by leveraging past experience. However, in the offline RL setting -- where a fixed collection of interactions are provided and no further interactions are allowed -- it has been shown that standard off-policy RL methods can significantly underperform. Recently proposed methods often aim to address this shortcoming by constraining learned policies to remain close to the given dataset of interactions. In this work, we closely investigate an important simplification of BCQ~\citep{fujimoto2018off} -- a prior approach for offline RL -- which removes a heuristic design choice and naturally restrict extracted policies to remain \emph{exactly} within the support of a given behavior policy. Importantly, in contrast to their original theoretical considerations, we derive this simplified algorithm through the introduction of a novel backup operator, Expected-Max Q-Learning (EMaQ), which is more closely related to the resulting practical algorithm. Specifically, in addition to the distribution support, EMaQ explicitly considers the number of samples and the proposal distribution, allowing us to derive new sub-optimality bounds which can serve as a novel measure of complexity for offline RL problems. In the offline RL setting -- the main focus of this work -- EMaQ matches and outperforms prior state-of-the-art in the D4RL benchmarks~\citep{fu2020d4rl}. In the online RL setting, we demonstrate that EMaQ is competitive with Soft Actor Critic (SAC). The key contributions of our empirical findings are demonstrating the importance of careful generative model design for estimating behavior policies, and an intuitive notion of complexity for offline RL problems. With its simple interpretation and fewer moving parts, such as no explicit function approximator representing the policy, EMaQ serves as a strong yet easy to implement baseline for future work.

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)0.005
metaresearch head score (Gemma)0.018
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.012
Threshold uncertainty score0.040

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0050.018
Meta-epidemiology (narrow)0.0020.001
Meta-epidemiology (broad)0.0020.001
Bibliometrics0.0010.001
Science and technology studies0.0010.002
Scholarly communication0.0020.002
Open science0.0040.004
Research integrity0.0030.005
Insufficient payload (model declined to judge)0.0120.003

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 head0.028
GPT teacher head0.320
Teacher spread0.291 · 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 designSimulation or modeling
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

Citations22
Published2021
Admission routes1
Has abstractyes

Explore more

Same venueInternational Conference on Machine LearningSame topicReinforcement Learning in RoboticsFrench-language works237,207