AI Paper Atlas

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

Experience Selection in Deep Reinforcement Learning for Control

Tim de Bruin, Jens Kober, Karl Tuyls, Robert Babuška

JMLR 2018 · Reinforcement learning › Policy gradient and actor-critic · approach: Regularized scalable deep value learning · 56 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
sample with PER instead of uniform samplingTDE(1.0) retention · Pendulum swing-up with noisemean absolute noise (position) in mini-batch: 1.713·10−2 -> 1.846·10−2Hurt
sample with PER instead of uniform samplingExpl(1.0) retention · Pendulum swing-up with noisemean absolute noise (position) in mini-batch: 1.584 ·10−2 -> 1.654 ·10−2Hurt
retain all experiences (FULL DB) with PER instead of FIFO with uniform samplingDDPG-style actor-critic · Pendulum swing-upnormalized mean rewardImproved
retain all experiences (FULL DB) with PER instead of FIFO with uniform samplingDDPG-style actor-critic · Magmannormalized mean rewardHurt
add PER sampling when sampling from diverse experiencesFIFO / Resv retention · Magmanfinal performanceHurt
add PER samplingFIFO retention · Magmanlearning speedImproved
add PER samplingretention strategies · Pendulum swing-uplearning speedImproved
add PER sampling under sensor and actuator noiseretention strategies · Pendulum and Magman with noise 0.02final performanceHurt
use reservoir retention instead of storing all experiencesResv[Uniform] with 10^4 buffer · Pendulum swing-upnormalized mean rewardNo clear effect
use a small FIFO buffer of recent experiencesFIFO[Uniform] · Magman at 50 Hzfinal performanceImproved
use FIFO retention at higher sampling frequency (200 Hz)FIFO[Uniform] · Magmanfinal performanceHurt
add sensor and actuator noiseexperience diversity need · Pendulum and Magmanneed for diverse dataNo clear effect

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