Mike Preuss
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Mike Preuss has a total of 221 co-authors. Here are the top co-authors by number of shared publications:
- Alan Kai Hassen
Leiden Institute of Advanced Computer Science, Leiden University, Leiden, The Netherlands
8 shared publications - Djork-Arné Clevert
Machine Learning & Computational Sciences, Pfizer Worldwide R&D, Berlin, Germany
7 shared publications - Christian Grimme
Department of Information Systems, University of Münster, Münster, Germany
6 shared publications - Samuel Genheden
Molecular AI, Discovery Sciences, R&D, AstraZeneca, Gothenburg, Sweden
6 shared publications - Sebastian Risi
IT University of Copenhagen, Copenhagen, Denmark
5 shared publications - Simon Wessing
Technische Universität Dortmund
5 shared publications
Latest Publications
- Open Access
- 11 September 2026
- Published by American Chemical Society (ACS) in ChemRxiv
Assessing the performance of implicit ensembling of neural networks for proteochemometric modeling with RandomNets
RandomNets, a recently introduced neural-network architecture that emulates the feature and sample bagging of Random Forests through an implicit, vectorised ensemble, has shown promising performance for quantitative structure–activity relationship (QSAR) modelling, but so far only on random data splits. Here we evaluate RandomNets in a proteochemometric (PCM) setting on two benchmarks: an original dataset (BtH) and a newly prepared dataset derived from ChEMBL 35, comprising 571,197 interactions across 382,493 unique compounds and 2,023 unique targets. Models were built from Morgan fingerprints together with additional physicochemical and protein descriptors, and assessed under both random and temporal splits. All models were trained within QSPRpred, enabling reproducible and benchmark-agnostic comparison against the best-performing PCM models from BtH, a deep neural network (DNN_PCM) and a Random Forest (RF_PCM). RandomNets with two layers gave the best overall performance, notably improving on the Matthews correlation coefficient (MCC) of the DNN baseline for the temporal split by 5.55% on the original dataset and 8.73% on the new dataset, while taking 2× less time to train on the original dataset and the same time on the new dataset. These results affirm the potential of RandomNets and extend it to PCM modelling.
- 11 September 2026
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessAssessing the performance of implicit ensembling of neural networks for proteochemometric modeling with RandomNets
RandomNets, a recently introduced neural-network architecture that emulates the feature and sample bagging of Random Forests through an implicit, vectorised ensemble, has shown promising performance for quantitative structure–activity relationship (QSAR) modelling, but so far only on random data splits. Here we evaluate RandomNets in a proteochemometric (PCM) setting on two benchmarks: an original dataset (BtH) and a newly prepared dataset derived from ChEMBL 35, comprising 571,197 interactions across 382,493 unique compounds and 2,023 unique targets. Models were built from Morgan fingerprints together with additional physicochemical and protein descriptors, and assessed under both random and temporal splits. All models were trained within QSPRpred, enabling reproducible and benchmark-agnostic comparison against the best-performing PCM models from BtH, a deep neural network (DNN_PCM) and a Random Forest (RF_PCM). RandomNets with two layers gave the best overall performance, notably improving on the Matthews correlation coefficient (MCC) of the DNN baseline for the temporal split by 5.55% on the original dataset and 8.73% on the new dataset, while taking 2× less time to train on the original dataset and the same time on the new dataset. These results affirm the potential of RandomNets and extend it to PCM modelling.
- 29 May 2026
- Published by MIT Press in Evolutionary Computation
- Vol. 34 (2) , 213-233
Illuminating the Diversity-Fitness Trade-Off in Black-Box Optimization
In real-world applications, users often favor structurally diverse design choices over one high-quality solution. It is therefore important to consider more solutions that decision makers can compare and further explore based on additional criteria. Alongside the existing approaches of evolutionary diversity optimization, quality diversity, and multimodal optimization, this paper presents a fresh perspective on this challenge by considering the problem of identifying a fixed number of solutions with a pairwise distance above a specified threshold while maximizing their average quality. We obtain first insight into these objectives by performing a subset selection on the search trajectories of different well-established search heuristics, whether they have been specifically designed with diversity in mind or not. We emphasize that the main goal of our work is not to present a new algorithm but to understand the capability of off-the-shelf algorithms to quantify the trade-off between the minimum pairwise distance within batches of solutions and their average quality. We also analyze how this trade-off depends on the properties of the underlying optimization problem. A possibly surprising outcome of our empirical study is the observation that naive uniform random sampling establishes a very strong baseline for our problem, hardly ever outperformed by the search trajectories of the considered heuristics. We interpret these results as a motivation to develop algorithms tailored to produce diverse solutions of high average quality.
- 29 May 2026
- Published by MIT Press in Evolutionary Computation
- Vol. 34 (2) , 213-233
Illuminating the Diversity-Fitness Trade-Off in Black-Box Optimization
In real-world applications, users often favor structurally diverse design choices over one high-quality solution. It is therefore important to consider more solutions that decision makers can compare and further explore based on additional criteria. Alongside the existing approaches of evolutionary diversity optimization, quality diversity, and multimodal optimization, this paper presents a fresh perspective on this challenge by considering the problem of identifying a fixed number of solutions with a pairwise distance above a specified threshold while maximizing their average quality. We obtain first insight into these objectives by performing a subset selection on the search trajectories of different well-established search heuristics, whether they have been specifically designed with diversity in mind or not. We emphasize that the main goal of our work is not to present a new algorithm but to understand the capability of off-the-shelf algorithms to quantify the trade-off between the minimum pairwise distance within batches of solutions and their average quality. We also analyze how this trade-off depends on the properties of the underlying optimization problem. A possibly surprising outcome of our empirical study is the observation that naive uniform random sampling establishes a very strong baseline for our problem, hardly ever outperformed by the search trajectories of the considered heuristics. We interpret these results as a motivation to develop algorithms tailored to produce diverse solutions of high average quality.
- Open Access
- 25 February 2026
- Published by Royal Society of Chemistry (RSC) in Digital Discovery
- Vol. 5 (4) , 1623-1634
Synthesis planning in reaction space: a study on success, robustness and diversity
Reaction-centric synthesis planning reduces search space redundancy. Online search outperforms self-play and is more robust to stock changes. Route diversity saturates across algorithms, making the single-step model, not search, the key limitation.
- 25 February 2026
- Published by Royal Society of Chemistry (RSC) in Digital Discovery
- Vol. 5 (4) , 1623-1634
Open AccessSynthesis planning in reaction space: a study on success, robustness and diversity
Reaction-centric synthesis planning reduces search space redundancy. Online search outperforms self-play and is more robust to stock changes. Route diversity saturates across algorithms, making the single-step model, not search, the key limitation.
- Top Cited Papers
- 30 December 2025
- Published by AI Access Foundation in Journal of Artificial Intelligence Research
- Vol. 84
Agentic Large Language Models, a Survey
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 agenda. Methods: Agentic LLMs are LLMs that (1) reason, (2) act, and (3) interact. We organize the literature according to these three categories. Results: The research in the first category focuses on reasoning, reflection, and retrieval, aiming to improve decision making; the second category focuses on action models, robots, and tools, aiming for agents that act as useful assistants; the third category focuses on multi-agent systems, aiming for collaborative task solving and simulating interaction to study emergent social behavior. We find that works mutually benefit from results in other categories: retrieval enables tool use, reflection improves multi-agent collaboration, and reasoning benefits all categories. Conclusions: We discuss applications of agentic LLMs and provide an agenda for further research. Important applications are in medical diagnosis, logistics and financial market analysis. Meanwhile, self-reflective agents playing roles and interacting with one another augment the process of scientific research itself. Further, agentic LLMs provide a solution for the problem of LLMs running out of training data: inference-time behavior generates new training states, such that LLMs can keep learning without needing ever larger datasets. We note that there is risk associated with LLM assistants taking action in the real world—safety, liability and security are open problems—while agentic LLMs are also likely to benefit society.
- 30 December 2025
- Published by AI Access Foundation in Journal of Artificial Intelligence Research
- Vol. 84
Top Cited PapersAgentic Large Language Models, a Survey
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 agenda. Methods: Agentic LLMs are LLMs that (1) reason, (2) act, and (3) interact. We organize the literature according to these three categories. Results: The research in the first category focuses on reasoning, reflection, and retrieval, aiming to improve decision making; the second category focuses on action models, robots, and tools, aiming for agents that act as useful assistants; the third category focuses on multi-agent systems, aiming for collaborative task solving and simulating interaction to study emergent social behavior. We find that works mutually benefit from results in other categories: retrieval enables tool use, reflection improves multi-agent collaboration, and reasoning benefits all categories. Conclusions: We discuss applications of agentic LLMs and provide an agenda for further research. Important applications are in medical diagnosis, logistics and financial market analysis. Meanwhile, self-reflective agents playing roles and interacting with one another augment the process of scientific research itself. Further, agentic LLMs provide a solution for the problem of LLMs running out of training data: inference-time behavior generates new training states, such that LLMs can keep learning without needing ever larger datasets. We note that there is risk associated with LLM assistants taking action in the real world—safety, liability and security are open problems—while agentic LLMs are also likely to benefit society.
- 16 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Evolutionary Computation
- Vol. PP (1089778X) , 1
An Evolution Strategy With Adaptive Fitness Sharing for Multimodal Multiobjective Optimization
Unlike multiobjective optimization (MOO), multimodal multiobjective optimization (MMMOO) should approximate the entire Pareto set, even if a portion of it maps onto the entire Pareto front. This study introduces a novel evolution strategy with adaptive fitness sharing (AFS) for MMMOO. The method, called, AFS-MMMO-ES, calculates an overall fitness for each solution in the selection pool based on its rank-wise hypervolume contribution, Pareto rank, and niche count in the decision space. Since the optimal niche radius is problem-dependent, this study introduces a novel strategy for on-the-fly adaptation of the niche radius. Simulations on meticulously designed test problems are performed to confirm the efficacy and reliability of this strategy in learning the optimal niche radius, as well as its significant impact on enhancing robustness and performance. Furthermore, AFS-MMMO-ES can easily reflect the relative importance of decision space diversity based on the decision-maker’s preference, a practically important feature that has been overlooked in this research field. Finally, the performance of AFS-MMMO-ES is assessed and compared with several successful MMMOO methods on a widely accepted test suite for MMMOO. Comparisons of numerical results reveal the robustness and superiority of AFS-MMMO-ES over its competitors.
- 16 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Evolutionary Computation
- Vol. PP (1089778X) , 1
An Evolution Strategy With Adaptive Fitness Sharing for Multimodal Multiobjective Optimization
Unlike multiobjective optimization (MOO), multimodal multiobjective optimization (MMMOO) should approximate the entire Pareto set, even if a portion of it maps onto the entire Pareto front. This study introduces a novel evolution strategy with adaptive fitness sharing (AFS) for MMMOO. The method, called, AFS-MMMO-ES, calculates an overall fitness for each solution in the selection pool based on its rank-wise hypervolume contribution, Pareto rank, and niche count in the decision space. Since the optimal niche radius is problem-dependent, this study introduces a novel strategy for on-the-fly adaptation of the niche radius. Simulations on meticulously designed test problems are performed to confirm the efficacy and reliability of this strategy in learning the optimal niche radius, as well as its significant impact on enhancing robustness and performance. Furthermore, AFS-MMMO-ES can easily reflect the relative importance of decision space diversity based on the decision-maker’s preference, a practically important feature that has been overlooked in this research field. Finally, the performance of AFS-MMMO-ES is assessed and compared with several successful MMMOO methods on a widely accepted test suite for MMMOO. Comparisons of numerical results reveal the robustness and superiority of AFS-MMMO-ES over its competitors.
- 26 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-8
Quantum Checkers: The Development and Analysis of a Quantum Combinatorial Game
This paper develops and analyses a novel quantum combinatorial game: quantum checkers (named Cheqqers). The concepts of superposition, entanglement, measurements and interference from quantum mechanics are integrated into the game of checkers by adding new types of legal moves. The addition of these new rules is done gradually by introducing several levels of 'quantumness'. Quantum checkers provides a framework for interpolating between a known and solved classical game and a more complex quantum game, and serves as 1) a benchmark for AI players learning to play quantum games and 2) an interesting game for human players that allows them to build intuition for quantum phenomena. We provide the initial analysis on the complexity of this game using random agents and a Monte Carlo tree search agent.
- 26 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-8
Quantum Checkers: The Development and Analysis of a Quantum Combinatorial Game
This paper develops and analyses a novel quantum combinatorial game: quantum checkers (named Cheqqers). The concepts of superposition, entanglement, measurements and interference from quantum mechanics are integrated into the game of checkers by adding new types of legal moves. The addition of these new rules is done gradually by introducing several levels of 'quantumness'. Quantum checkers provides a framework for interpolating between a known and solved classical game and a more complex quantum game, and serves as 1) a benchmark for AI players learning to play quantum games and 2) an interesting game for human players that allows them to build intuition for quantum phenomena. We provide the initial analysis on the complexity of this game using random agents and a Monte Carlo tree search agent.
- Open Access
- 14 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 1900-1908
Cascading CMA-ES Instances for Generating Input-diverse Solution Batches
- 14 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 1900-1908
Open AccessCascading CMA-ES Instances for Generating Input-diverse Solution Batches
- Open Access
- 30 June 2025
- Published by American Chemical Society (ACS) in ChemRxiv
Synthesis Planning in Reaction Space: A Study on Success, Robustness and Diversity
Computer-aided synthesis planning aims to identify viable synthetic routes from a target compound to readily available building blocks by iteratively decomposing molecules into smaller precursors. Self-play search algorithms, trained with simulated experience, reach state-of-the-art performance. However, these methods typically plan in the molecular rather than the reaction space, leading to redundant or near-duplicate reaction outcomes in the search tree. In this work, we introduce a reaction-centric planning approach that measures the novelty of proposed reactions, thereby constraining the search problem to genuinely unexplored disconnection ideas, i.e., unique ways of decomposing molecules using reactions. Our results show that the overall synthesis planning search space is much smaller than expected due to the absence of diverse disconnection ideas within the underlying retrosynthesis neural network. Surprisingly, we also find that, under a reasonable time budget of less than an hour, online search algorithms outperform state-of-the-art self-play methods and are more robust to environmental changes, such as minor modifications to the available purchasable building blocks.
- 30 June 2025
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessSynthesis Planning in Reaction Space: A Study on Success, Robustness and Diversity
Computer-aided synthesis planning aims to identify viable synthetic routes from a target compound to readily available building blocks by iteratively decomposing molecules into smaller precursors. Self-play search algorithms, trained with simulated experience, reach state-of-the-art performance. However, these methods typically plan in the molecular rather than the reaction space, leading to redundant or near-duplicate reaction outcomes in the search tree. In this work, we introduce a reaction-centric planning approach that measures the novelty of proposed reactions, thereby constraining the search problem to genuinely unexplored disconnection ideas, i.e., unique ways of decomposing molecules using reactions. Our results show that the overall synthesis planning search space is much smaller than expected due to the absence of diverse disconnection ideas within the underlying retrosynthesis neural network. Surprisingly, we also find that, under a reasonable time budget of less than an hour, online search algorithms outperform state-of-the-art self-play methods and are more robust to environmental changes, such as minor modifications to the available purchasable building blocks.
- Open Access
- 15 April 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-14
Towards a Celeste AI Framework: Agent-free Automated 2D Level Generation for Multidirectional Platformers
- 15 April 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-14
Open AccessTowards a Celeste AI Framework: Agent-free Automated 2D Level Generation for Multidirectional Platformers
- Open Access
- 28 March 2025
- Published by Springer Nature in Journal of Cheminformatics
- Vol. 17 (1) , 1-16
Generate what you can make: achieving in-house synthesizability with readily available resources in de novo drug design
Computer-Aided Synthesis Planning (CASP) and CASP-based approximated synthesizability scores have rarely been used as generation objectives in Computer-Aided Drug Design despite facilitating the in-silico generation of synthesizable molecules. However, these synthesizability approaches are disconnected from the reality of small laboratory drug design, where building block resources are limited, thus making the notion of in-house synthesizability with already available resources highly desirable. In this work, we show a successful in-house de novo drug design workflow generating active and in-house synthesizable ligands of monoglyceride lipase (MGLL). First, we demonstrate the successful transfer of CASP from 17.4 million commercial building blocks to a small laboratory setting of roughly 6000 building blocks with only a decrease of –12% in CASP success when accepting two reaction-steps longer synthesis routes on average. Next, we present a rapidly retrainable in-house synthesizability score, successfully capturing our in-house synthesizability without relying on external building block resources. We show that including our in-house synthesizability score in a multi-objective de novo drug design workflow, alongside a simple QSAR model, provides thousands of potentially active and easily in-house synthesizable molecules. Finally, we experimentally evaluate the synthesis and biochemical activity of three de novo candidates using their CASP-suggested synthesis routes employing only in-house building blocks. We find one candidate with evident activity, suggesting potential new ligand ideas for MGLL inhibitors while showcasing the usefulness of our in-house synthesizability score for de novo drug design. Scientific contribution Our core scientific contribution is the introduction of in-house de novo drug design, which enables the practical application of generative methods in small laboratories by utilizing a limited stock of available building blocks. Our fast-to-adapt workflow for in-house synthesizability scoring requires minimal computational retraining costs while supporting a high diversity of generated structures. We highlight the practicality of our approach through a comprehensive in-vitro case study that relies entirely on in-house resources, including in-silico generation, synthesis planning, and activity evaluation.
- 28 March 2025
- Published by Springer Nature in Journal of Cheminformatics
- Vol. 17 (1) , 1-16
Open AccessGenerate what you can make: achieving in-house synthesizability with readily available resources in de novo drug design
Computer-Aided Synthesis Planning (CASP) and CASP-based approximated synthesizability scores have rarely been used as generation objectives in Computer-Aided Drug Design despite facilitating the in-silico generation of synthesizable molecules. However, these synthesizability approaches are disconnected from the reality of small laboratory drug design, where building block resources are limited, thus making the notion of in-house synthesizability with already available resources highly desirable. In this work, we show a successful in-house de novo drug design workflow generating active and in-house synthesizable ligands of monoglyceride lipase (MGLL). First, we demonstrate the successful transfer of CASP from 17.4 million commercial building blocks to a small laboratory setting of roughly 6000 building blocks with only a decrease of –12% in CASP success when accepting two reaction-steps longer synthesis routes on average. Next, we present a rapidly retrainable in-house synthesizability score, successfully capturing our in-house synthesizability without relying on external building block resources. We show that including our in-house synthesizability score in a multi-objective de novo drug design workflow, alongside a simple QSAR model, provides thousands of potentially active and easily in-house synthesizable molecules. Finally, we experimentally evaluate the synthesis and biochemical activity of three de novo candidates using their CASP-suggested synthesis routes employing only in-house building blocks. We find one candidate with evident activity, suggesting potential new ligand ideas for MGLL inhibitors while showcasing the usefulness of our in-house synthesizability score for de novo drug design. Scientific contribution Our core scientific contribution is the introduction of in-house de novo drug design, which enables the practical application of generative methods in small laboratories by utilizing a limited stock of available building blocks. Our fast-to-adapt workflow for in-house synthesizability scoring requires minimal computational retraining costs while supporting a high diversity of generated structures. We highlight the practicality of our approach through a comprehensive in-vitro case study that relies entirely on in-house resources, including in-silico generation, synthesis planning, and activity evaluation.