Comments (2)
From my understanding there are two purposes of the entropy regularization in max ent RL. The first purpose is to improve exploration, and to achieve this you need to back up the entropy in the Bellman backup (--cql.backup_entropy
option). However in offline RL we don't need to explore at all, and this is exactly why we turn off that option by default. This brings us to the second purpose of max entropy, which is improve the policy optimization landscape. Entropy regularization prevents the policy action distribution from collapsing to a single point and this helps even in offline RL, since we need to query the policy's actions when computing next step value. This is the reason why we also have it when doing behavior cloning in the beginning.
from jaxcql.
Many thanks for your kind reply.
from jaxcql.
Related Issues (4)
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from jaxcql.