technique

Variance Reduction

techniqueactivevariance-reduction-61ff8044·1 events·first seen 28d ago

Aliases: Variance Reduction

Co-occurring entities

Action-Dependent Factorized Baselines Policy Gradient Methods OpenAI

More like this (12)

ELBO variance minimization posterior predictive variance minimization CVaR (Conditional Value at Risk)Invariant Risk Minimization adversarial refinement Uncertainty Calibration Expected Improvement regulatory variant-effect prediction Remote VAE Divide-and-Conquer Value Learning R-Drop consistency regularization supply chain re-optimization

Recent events (1)

3Openai Blog·28d ago·source ↗

Variance Reduction for Policy Gradient with Action-Dependent Factorized Baselines

OpenAI published a research paper on variance reduction techniques for policy gradient methods in reinforcement learning. The work introduces action-dependent factorized baselines as a way to reduce variance in policy gradient estimates without introducing bias. This is a foundational RL training methodology contribution relevant to improving sample efficiency in reinforcement learning.

Alignment and RLHF Action-Dependent Factorized Baselines Policy Gradient Methods Variance Reduction +1 more