AI Paper Atlas

Research map / Reinforcement learning / Policy gradient and actor-critic

Gaussian Approximation for Bias Reduction in Q-Learning

Carlo D'Eramo, Andrea Cini, Alessandro Nuara, Matteo Pirotta, Cesare Alippi, Jan Peters, Marcello Restelli

JMLR 2021 · Reinforcement learning › Policy gradient and actor-critic · approach: Temporal-difference and discounting theory · 6 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace the Maximum Estimator with the Weighted Estimator in absolute-bias comparison of two Gaussian variablesWeighted Estimator · Two Gaussians, sigma=10, 100 samples eachabsolute bias: between 0.5 and 1.8Mixed
use Weighted Estimator instead of Double Estimator when means differ by littleWeighted Estimator · Two Gaussians, sigma=10, 100 samples eachabsolute bias: less than 0.5Hurt
use WE instead of ME, DE, MME in internet ads MABWeighted Estimator · Internet ads (CTR estimation)MSEImproved
use WE instead of ME in sponsored search auctionsWeighted Estimator · Sponsored search auctionsincentive compatibilityImproved
use Double Estimator instead of ME in sponsored search auctionsDouble Estimator · Sponsored search auctionsincentive compatibilityMixed
add weighted policy (Thompson-like sampling) to Weighted Q-Learning (ablation)Weighted Q-Learning · Grid world (Gaussian sigma=1)average reward per stepImproved
use Weighted Q-Learning vs ME, DE, MME on noisy grid worldWeighted Q-Learning · Grid world (noisy rewards)learning speed of best policyHurt
use WE with infinite actions (GP) instead of binned ME/DE/MMEWE infinity · Pricing problembiasImproved
use Weighted FQI instead of other FQI variants with discrete actionsWeighted FQI · Swing-Up Pendulum (11 discrete actions)average reward: less or equal to 50Improved
use continuous-action WE (infinity) in FQIWeighted FQI · Swing-Up Pendulum (continuous actions)average rewardNo clear effect
use WDQN instead of DDQN and AvgDQNWDQN · Lunar Lander (stochastic)sample efficiency and prediction accuracyImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy