Explainable Reinforcement Learning for Longitudinal Control

Photo by James Pond on Unsplash

The following article, presents the research which I conducted together with Jan Dohmen and Marco Wiering.

TL;DR: Reinforcement Learning is showing promise in achieving optimal performances in numerous applications. However, as long as the learned actions remain intransparent, their use in security-relevant applications is unlikely. The new RL-SHAP Diagram presented here opens the black box and gives a new perspective to the Reinforcement Learning decision-making.

Story line

Imagine you are an engineer working on self-driving vehicles. An exciting idea, isn’t it?

It gets even better. You spent time persuading your boss to use machine learning for longitudinal control, he agreed!

You started…


Reinforcement Learning optimization under uncertainties.

Here I present research with Lucas Vogt, Jan Dohmen and Christoph Friebel.

TL;DR: Controls for technical systems can be optimized in the simulation. In reality, however, numerous unknowns are waiting for us. In this post, we show how the addition of noise and sensor errors affects the optimization result of a Reinforcement Learning agent.

Photo by Kyle Glenn on Unsplash

Motivation

“Better safe than sorry!”

The most car drivers are following this idea because in the long run, it is more advantageous to sometimes obtain a suboptimal result than to push for an optimal result every time. For example, motorists seldom push the performance of their vehicle…


Just try our new LongiControl Environment

Photo by NeONBRAND on Unsplash

Here I present research with Jan Dohmen and Christoph Friebel.

Motivation

Recent years have seen a surge of applicative successes using Reinforcement Learning (RL) [1] to solve challenging games and smaller domain problems [2][3][4]. These successes in RL have been achieved in part due to the strong collaborative effort by the RL community to work on common, open-sourced environment simulators such as OpenAI’s Gym [5] that allow for expedited development and valid comparisons between different, state-of-art strategies.

However many existing environments contain games rather than real-world problems. Only recent publications initiate the transition to application-oriented RL [6][7]. In this contribution, we…

Roman Liessner

Hi, I’m Roman, an Engineer with a PhD in Automotive Engineering and a passion for Artificial Intelligence.

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