Publications (41)
On the Verge of Solving Rocket League using Deep Reinforcement Learning and Sim-to-sim Transfer
Marco Pleines, Konstantin Ramthun, Yannik Wegener +12
Autonomously trained agents that are supposed to play video games reasonably well rely either on fast simulation speeds or heavy parallelization across thousands of machines runnin…
Social Bots: Human-Like by Means of Human Control?
Christian Grimme, Mike Preuss, Lena Adam +1
Social bots are currently regarded an influential but also somewhat mysterious factor in public discourse and opinion making. They are considered to be capable of massively distrib…
From Chess and Atari to StarCraft and Beyond: How Game AI is Driving the World of AI
Sebastian Risi, Mike Preuss
This paper reviews the field of Game AI, which not only deals with creating agents that can play a certain game, but also with areas as diverse as creating game content automatical…
Mind the Retrosynthesis Gap: Bridging the divide between Single-step and Multi-step Retrosynthesis Prediction
Alan Kai Hassen, Paula Torren-Peraire, Samuel Genheden +3
Retrosynthesis is the task of breaking down a chemical compound recursively step-by-step into molecular precursors until a set of commercially available molecules is found. Consequ…
When to Go, and When to Explore: The Benefit of Post-Exploration in Intrinsic Motivation
Zhao Yang, Thomas M. Moerland, Mike Preuss +1
Go-Explore achieved breakthrough performance on challenging reinforcement learning (RL) tasks with sparse rewards. The key insight of Go-Explore was that successful exploration req…
An Analysis of Phenotypic Diversity in Multi-Solution Optimization
Alexander Hagg, Mike Preuss, Alexander Asteroth +1
More and more, optimization methods are used to find diverse solution sets. We compare solution diversity in multi-objective optimization, multimodal optimization, and quality dive…
The True Destination of EGO is Multi-local Optimization
Simon Wessing, Mike Preuss
Efficient global optimization is a popular algorithm for the optimization of expensive multimodal black-box functions. One important reason for its popularity is its theoretical fo…
Believable Minecraft Settlements by Means of Decentralised Iterative Planning
Arthur van der Staaij, Jelmer Prins, Vincent L. Prins +4
Procedural city generation that focuses on believability and adaptability to random terrain is a difficult challenge in the field of Procedural Content Generation (PCG). Dozens of…
Versatile Black-Box Optimization
Jialin Liu, Antoine Moreau, Mike Preuss +4
Choosing automatically the right algorithm using problem descriptors is a classical component of combinatorial optimization. It is also a good tool for making evolutionary algorith…
In Trust We Survive: Emergent Trust Learning
Qianpu Chen, Giulio Barbero, Mike Preuss +1
We introduce Emergent Trust Learning (ETL), a lightweight, trust-based control algorithm that can be plugged into existing AI agents. It enables these to reach cooperation in compe…
Warm-Start AlphaZero Self-Play Search Enhancements
Hui Wang, Mike Preuss, Aske Plaat
Recently, AlphaZero has achieved landmark results in deep reinforcement learning, by providing a single self-play architecture that learned three different games at super human lev…
Atom-anchored LLMs speak Chemistry: A Retrosynthesis Demonstration
Alan Kai Hassen, Andrius Bernatavicius, Antonius P. A. Janssen +3
Applications of machine learning in chemistry are often limited by the scarcity and expense of labeled data, restricting traditional supervised methods. In this work, we introduce…
Generalization, Mayhems and Limits in Recurrent Proximal Policy Optimization
Marco Pleines, Matthias Pallasch, Frank Zimmer +1
At first sight it may seem straightforward to use recurrent layers in Deep Reinforcement Learning algorithms to enable agents to make use of memory in the setting of partially obse…
Analysis of Hyper-Parameters for Small Games: Iterations or Epochs in Self-Play?
Hui Wang, Michael Emmerich, Mike Preuss +1
The landmark achievements of AlphaGo Zero have created great research interest into self-play in reinforcement learning. In self-play, Monte Carlo Tree Search is used to train a de…
Impressions of the GDMC AI Settlement Generation Challenge in Minecraft
Christoph Salge, Claus Aranha, Adrian Brightmoore +15
The GDMC AI settlement generation challenge is a PCG competition about producing an algorithm that can create an "interesting" Minecraft settlement for a given map. This paper cont…
Cascading CMA-ES Instances for Generating Input-diverse Solution Batches
Maria Laura Santoni, Christoph Dürr, Carola Doerr +2
Rather than obtaining a single good solution for a given optimization problem, users often seek alternative design choices, because the best-found solution may perform poorly with…
A New Challenge: Approaching Tetris Link with AI
Matthias Muller-Brockhausen, Mike Preuss, Aske Plaat
Decades of research have been invested in making computer programs for playing games such as Chess and Go. This paper focuses on a new game, Tetris Link, a board game that is still…
Agentic Large Language Models, a survey
Aske Plaat, Max van Duijn, Niki van Stein +3
Background: There is great interest in agentic LLMs, large language models that act as agents. Objectives: We review the growing body of work in this area and provide a research ag…
Tackling Morpion Solitaire with AlphaZero-likeRanked Reward Reinforcement Learning
Hui Wang, Mike Preuss, Michael Emmerich +1
Morpion Solitaire is a popular single player game, performed with paper and pencil. Due to its large state space (on the order of the game of Go) traditional search algorithms, suc…
Continuous Episodic Control
Zhao Yang, Thomas M. Moerland, Mike Preuss +1
Non-parametric episodic memory can be used to quickly latch onto high-rewarded experience in reinforcement learning tasks. In contrast to parametric deep reinforcement learning app…
Procedural Content Generation: Better Benchmarks for Transfer Reinforcement Learning
Matthias Müller-Brockhausen, Mike Preuss, Aske Plaat
The idea of transfer in reinforcement learning (TRL) is intriguing: being able to transfer knowledge from one problem to another problem without learning everything from scratch. T…
Deep Model-Based Reinforcement Learning for High-Dimensional Problems, a Survey
Aske Plaat, Walter Kosters, Mike Preuss
Deep reinforcement learning has shown remarkable success in the past few years. Highly complex sequential decision making problems have been solved in tasks such as game playing an…
First Go, then Post-Explore: the Benefits of Post-Exploration in Intrinsic Motivation
Zhao Yang, Thomas M. Moerland, Mike Preuss +1
Go-Explore achieved breakthrough performance on challenging reinforcement learning (RL) tasks with sparse rewards. The key insight of Go-Explore was that successful exploration req…
Adaptive Warm-Start MCTS in AlphaZero-like Deep Reinforcement Learning
Hui Wang, Mike Preuss, Aske Plaat
AlphaZero has achieved impressive performance in deep reinforcement learning by utilizing an architecture that combines search and training of a neural network in self-play. Many r…
Hyper-Parameter Sweep on AlphaZero General
Hui Wang, Michael Emmerich, Mike Preuss +1
Since AlphaGo and AlphaGo Zero have achieved breakground successes in the game of Go, the programs have been generalized to solve other tasks. Subsequently, AlphaZero was developed…
Learning to Plan Chemical Syntheses
Marwin H. S. Segler, Mike Preuss, Mark P. Waller
From medicines to materials, small organic molecules are indispensable for human well-being. To plan their syntheses, chemists employ a problem solving technique called retrosynthe…
Models Matter: The Impact of Single-Step Retrosynthesis on Synthesis Planning
Paula Torren-Peraire, Alan Kai Hassen, Samuel Genheden +4
Retrosynthesis consists of breaking down a chemical compound recursively step-by-step into molecular precursors until a set of commercially available molecules is found with the go…
High-Accuracy Model-Based Reinforcement Learning, a Survey
Aske Plaat, Walter Kosters, Mike Preuss
Deep reinforcement learning has shown remarkable success in the past few years. Highly complex sequential decision making problems from game playing and robotics have been solved w…
Quantum Checkers: The Development and Analysis of a Quantum Combinatorial Game
Marien Raat, Luuk van den Nouweland, Matthias Müller-Brockhausen +2
This paper develops and analyses a novel quantum combinatorial game: quantum checkers (codenamed Cheqqers). The concepts of superposition, entanglement, measurements and interferen…
FakeYou! -- A Gamified Approach for Building and Evaluating Resilience Against Fake News
Lena Clever, Dennis Assenmacher, Kilian Müller +4
Nowadays fake news are heavily discussed in public and political debates. Even though the phenomenon of intended false information is rather old, misinformation reaches a new level…
Reliable validation of Reinforcement Learning Benchmarks
Matthias Müller-Brockhausen, Aske Plaat, Mike Preuss
Reinforcement Learning (RL) is one of the most dynamic research areas in Game AI and AI as a whole, and a wide variety of games are used as its prominent test problems. However, it…
Reset-free Reinforcement Learning with World Models
Zhao Yang, Thomas M. Moerland, Mike Preuss +2
Reinforcement learning (RL) is an appealing paradigm for training intelligent agents, enabling policy acquisition from the agent's own autonomously acquired experience. However, th…
Potential-based Reward Shaping in Sokoban
Zhao Yang, Mike Preuss, Aske Plaat
Learning to solve sparse-reward reinforcement learning problems is difficult, due to the lack of guidance towards the goal. But in some problems, prior knowledge can be used to aug…
Transfer Learning and Curriculum Learning in Sokoban
Zhao Yang, Mike Preuss, Aske Plaat
Transfer learning can speed up training in machine learning and is regularly used in classification tasks. It reuses prior knowledge from other tasks to pre-train networks for new…
Two-Memory Reinforcement Learning
Zhao Yang, Thomas. M. Moerland, Mike Preuss +1
While deep reinforcement learning has shown important empirical success, it tends to learn relatively slow due to slow propagation of rewards information and slow update of paramet…
Illuminating the Diversity-Fitness Trade-Off in Black-Box Optimization
Maria Laura Santoni, Elena Raponi, Aneta Neumann +3
In real-world applications, users often favor structurally diverse design choices over one high-quality solution. It is hence important to consider more solutions that decision mak…
Memory Gym: Towards Endless Tasks to Benchmark Memory Capabilities of Agents
Marco Pleines, Matthias Pallasch, Frank Zimmer +1
Memory Gym presents a suite of 2D partially observable environments, namely Mortar Mayhem, Mystery Path, and Searing Spotlights, designed to benchmark memory capabilities in decisi…
Applications of Artificial Intelligence in Live Action Role-Playing Games (LARP)
Christoph Salge, Emily Short, Mike Preuss +2
Live Action Role-Playing (LARP) games and similar experiences are becoming a popular game genre. Here, we discuss how artificial intelligence techniques, particularly those commonl…
Obstacle Tower Without Human Demonstrations: How Far a Deep Feed-Forward Network Goes with Reinforcement Learning
Marco Pleines, Jenia Jitsev, Mike Preuss +1
The Obstacle Tower Challenge is the task to master a procedurally generated chain of levels that subsequently get harder to complete. Whereas the most top performing entries of las…
Guiding Skill Discovery with Foundation Models
Zhao Yang, Thomas M. Moerland, Mike Preuss +3
Learning diverse skills without hand-crafted reward functions could accelerate reinforcement learning in downstream tasks. However, existing skill discovery methods focus solely on…
Pokemon Red via Reinforcement Learning
Marco Pleines, Daniel Addis, David Rubinstein +3
Pokémon Red, a classic Game Boy JRPG, presents significant challenges as a testbed for agents, including multi-tasking, long horizons of tens of thousands of steps, hard explorati…