nolisten / erl Goto Github PK
View Code? Open in Web Editor NEWExploration based Reinforcement Learning. (Montezuma Revenge)
License: MIT License
Exploration based Reinforcement Learning. (Montezuma Revenge)
License: MIT License
I found there might be a small problem in your way of sampling:
index = self.rng.randint(self.bottom,
self.bottom + self.size - self.phi_length)
When the number of experience exceeds self.size, the value of "self.bottom" will be incremented:
if self.size == self.max_steps:
self.bottom = (self.bottom + 1) % self.max_steps
But it seems that the sampling should start from index 0, instead of "self.bottom", to uniformly sample from the experience replay.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.