AI Paper Atlas

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

Managing Temporal Resolution in Continuous Value Estimation: A Fundamental Trade-off

Zichen Zhang, Johannes Kirschner, Junxi Zhang, Francesco Zanini, Alex Ayoub, Masood Dehghan, Dale Schuurmans

NeurIPS 2023 · Reinforcement learning › Policy gradient and actor-critic · approach: Temporal-difference and discounting theory · 0 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
vary time step-size h from small to large at fixed data budgetMonte-Carlo policy evaluation · Langevin / 1-D LQR (T=8, a=-1)MSE of value estimateMixed
increase data budget BMonte-Carlo policy evaluation · 1-D LQRMSEImproved
increase data budget B (optimal step-size)optimal step-size h* · 1-D LQRh*Improved
increase absolute drift coefficient aoptimal step-size h* · 1-D LQRh*Improved
change eigenvalues of A in 3x3 stable random matricesoptimal step-size h* · 3-D LQR finite and infinite horizonh*No clear effect
use a fixed pre-set step-size regardless of data budgetMonte-Carlo policy evaluation · LQR and MuJoCoMSEHurt
tune step-size h in nonlinear control systemsMonte-Carlo policy evaluation · Pendulum, BipedalWalker, six MuJoCo tasksMSEImproved
increase data budget B in nonlinear systemsoptimal step-size h* · MuJoCo and Gym tasksh*Mixed
set h* by the scaling h* = c_F B^(-1/3) fitted from a small budgetstep-size selection · nonlinear environmentsh* vs BMixed
use infinite-horizon scaling h* ~ c_I B^(-1/5) rather than finite-horizon B^(-1/3)step-size selection · discounted infinite-horizon LQRh* vs BMixed

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