This paper proposes a probabilistic planner that can solve goal-conditional tasks such as complex continuous control problems. The approach reaches state-of-the-art performance when compared to current deep reinforcement learning algorithms. However, the method relies on an ensemble of deep generative models and is computationally intensive. It would be interesting to reproduce the results presented in this paper on their robotic manipulation and navigation problems as these are very challenging problems that current reinforcement learning methods cannot easily solve (and when they do, they require a significantly larger number of experiences). Can the results be reproduced out-of-the-box with the provided code?
In theory, reproducing this paper should only require a clone of a public Git repository, and the execution of a Makefile (detailed in the README of the paper repository at https://github.com/psychoinformatics-de/paper-remodnav). We've set up our paper to be dynamically generated, retrieving and installing the relevant data and software automatically, and we've even created a tutorial about it, so that others can reuse the same setup for their work. Nevertheless, we've for example never tried it out across different operating systems - who knows whether it works on Windows? We'd love to share the tips and tricks we found to work, and even more love feedback on how to improve this further.
I guess it could be a cool learning experience. The paper is written with knitr, uses a seed, is part of the R package it describes, was openly written using version control (SVN, R-Forge) and is available in an open access journal (@up_jors).