Open in app
Jonathan Hui
17.2K Followers
About

Sign in

17.2K Followers
About
Open in app
Thank you for your post, a small question,
1

Xc Bao

Jonathan Hui

Jonathan Hui

Jan 16, 2019·1 min read

This will give you more details on importance sampling.

RL — Importance Sampling

Motivation in RL

medium.com

But, in short:

The variance depends on how you sample data. That is why you cannot use the old policy for too long for the practical purpose.

I don’t know what do you mean by “shrink”.

See Dec 9, 2018 for a previous answer that may relate to your question.

Written by

Jonathan Hui

Deep Learning

More from Jonathan Hui

Deep Learning

More From Medium

TensorFlow Save & Restore Model

Jonathan Hui

TensorFlow RNN models

Jonathan Hui

TensorFlow Dataset & Data Preparation

Jonathan Hui

TensorFlow Custom training, Transfer learning & Custom layers

Jonathan Hui

TensorFlow Eager Execution v.s. Graph (@tf.function)

Jonathan Hui

F is a second-order derivative.

Jonathan Hui

TensorFlow & Keras

Jonathan Hui

TensorFlow Automatic Differentiation (AutoDiff)

Jonathan Hui

About

Help

Legal

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store