Introduction If you have ever taken a class in mathematical optimization, you are probably familiar with the notion of a constrained optimization problem. If not, then don’t worry, because the idea is really simple.

Deep Q networks have proven to be an easy to implement method for solving control problems in both continuous or large discrete state spaces. The action-specific deep recurrent Q network (ADRQN) introduces an intermediate LSTM layer for remembering action-observation pairs when dealing with partial observability. This post explores a compact PyTorch implementation of the ADRQN including small scale experiments on classical control tasks.

Joint work with Cecilia Casolo and Mats van Tongeren. Published to https://reproducedpapers.org, a TU Delft repository for independent paper reproductions in Computer Vision and Deep Learning.

Published with Wowchemy — the free, open source website builder that empowers creators.