Jacob de Nobel
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jacob de Nobel has a total of 31 co-authors. Here are the top co-authors by number of shared publications:
- Thomas Bäck
Leiden University
17 shared publications - Diederick Vermetten
Sorbonne Université, CNRS, LIP6
14 shared publications - Anna V. Kononova
LIACS, Leiden University
8 shared publications - Furong Ye
Leiden Institute of Advanced Computer Science, Leiden University, Leiden, Netherlands
8 shared publications - Carola Doerr
Sorbonne Université, CNRS, LIP6
7 shared publications - Hao Wang
Leiden University
6 shared publications
Latest Publications
- 2 September 2026
- Published by Springer Nature
- p. 50-65
Integer Natural Evolution Strategies
- Open Access
- 23 July 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Evolutionary Learning and Optimization
Solving Deep Reinforcement Learning Tasks with Evolution Strategies and Linear Policy Networks
Although deep reinforcement learning methods can learn effective policies for challenging problems, the underlying algorithms are complex, and training times are often long. This study investigates how several state-of-the-art versions of Evolution Strategies perform compared to gradient-based deep reinforcement learning methods. We use Evolution Strategies to optimise the weights of a neural network via neuroevolution, performing direct policy search. We benchmark deep and linear policy networks for gradient-based methods, including Proximal Policy optimisation, Soft Actor-Critic, and Deep Q-Learning. These are evaluated against linear Evolution Strategies and Augmented Random Search, and additionally against scalable ES for deeper policies, such as LM-MA-ES. Our results reveal that Evolution Strategies can find effective linear policies for many reinforcement learning benchmark tasks, unlike deep reinforcement learning methods that can only find successful policies using much larger networks, suggesting that current benchmarks are easier to solve than previously assumed. Linear Evolution Strategies also outperform scalable ES variants trained on larger neural networks. Interestingly, Evolution Strategies, which do not use any gradient information, also achieve results comparable to gradient-based deep reinforcement learning algorithms for higher-complexity tasks. Furthermore, by directly accessing the memory state of the game, Evolution Strategies can find successful policies in Atari that outperform the policies found by Deep Q-Learning. Evolution Strategies also outperform Augmented Random Search in most benchmarks, demonstrating superior sample efficiency and robustness in training linear policy networks.
- 23 July 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Evolutionary Learning and Optimization
Open AccessSolving Deep Reinforcement Learning Tasks with Evolution Strategies and Linear Policy Networks
Although deep reinforcement learning methods can learn effective policies for challenging problems, the underlying algorithms are complex, and training times are often long. This study investigates how several state-of-the-art versions of Evolution Strategies perform compared to gradient-based deep reinforcement learning methods. We use Evolution Strategies to optimise the weights of a neural network via neuroevolution, performing direct policy search. We benchmark deep and linear policy networks for gradient-based methods, including Proximal Policy optimisation, Soft Actor-Critic, and Deep Q-Learning. These are evaluated against linear Evolution Strategies and Augmented Random Search, and additionally against scalable ES for deeper policies, such as LM-MA-ES. Our results reveal that Evolution Strategies can find effective linear policies for many reinforcement learning benchmark tasks, unlike deep reinforcement learning methods that can only find successful policies using much larger networks, suggesting that current benchmarks are easier to solve than previously assumed. Linear Evolution Strategies also outperform scalable ES variants trained on larger neural networks. Interestingly, Evolution Strategies, which do not use any gradient information, also achieve results comparable to gradient-based deep reinforcement learning algorithms for higher-complexity tasks. Furthermore, by directly accessing the memory state of the game, Evolution Strategies can find successful policies in Atari that outperform the policies found by Deep Q-Learning. Evolution Strategies also outperform Augmented Random Search in most benchmarks, demonstrating superior sample efficiency and robustness in training linear policy networks.
- Open Access
- 13 November 2025
- Published by Elsevier in Hearing Research
- Vol. 469, 109469
From spikes to speech: NeuroVoc — A biologically plausible vocoder framework for auditory perception and cochlear implant simulation
- 13 November 2025
- Published by Elsevier in Hearing Research
- Vol. 469, 109469
Open AccessFrom spikes to speech: NeuroVoc — A biologically plausible vocoder framework for auditory perception and cochlear implant simulation
- Open Access
- 14 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 1396-1407
Benchmarking single- and multi-objective optimization algorithms: how to make your experimental data more valuable
- 14 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 1396-1407
Open AccessBenchmarking single- and multi-objective optimization algorithms: how to make your experimental data more valuable
- Open Access
- 13 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 701-709
Abnormal Mutations: Evolution Strategies Don't Require Gaussianity
- 13 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 701-709
Open AccessAbnormal Mutations: Evolution Strategies Don't Require Gaussianity
- Open Access
Sampling in CMA-ES: Low Numbers of Low Discrepancy Points
The Covariance Matrix Adaptation Evolution Strategy (CMA-ES) is one of the most successful examples of a derandomized evolution strategy. However, it still relies on randomly sampling offspring, which can be done via a uniform distribution and subsequently transforming into the required Gaussian. Previous work has shown that replacing this uniform sampling with a low-discrepancy sampler, such as Halton or Sobol sequences, can improve performance over a wide set of problems. We show that iterating through small, fixed sets of low-discrepancy points can still perform better than the default uniform distribution. Moreover, using only 128 points throughout the search is sufficient to closely approximate the empirical performance of using the complete pseudorandom sequence up to dimensionality 40 on the BBOB benchmark. For lower dimensionalities (below 10), we find that using as little as 32 unique low discrepancy points performs similar or better than uniform sampling. In 2D, for which we have highly optimized low discrepancy samples available, we demonstrate that using these points yields the highest empirical performance and requires only 16 samples to improve over uniform sampling. Overall, we establish a clear relation between the $L_2$ discrepancy of the used point set and the empirical performance of the CMA-ES.Open AccessSampling in CMA-ES: Low Numbers of Low Discrepancy Points
The Covariance Matrix Adaptation Evolution Strategy (CMA-ES) is one of the most successful examples of a derandomized evolution strategy. However, it still relies on randomly sampling offspring, which can be done via a uniform distribution and subsequently transforming into the required Gaussian. Previous work has shown that replacing this uniform sampling with a low-discrepancy sampler, such as Halton or Sobol sequences, can improve performance over a wide set of problems. We show that iterating through small, fixed sets of low-discrepancy points can still perform better than the default uniform distribution. Moreover, using only 128 points throughout the search is sufficient to closely approximate the empirical performance of using the complete pseudorandom sequence up to dimensionality 40 on the BBOB benchmark. For lower dimensionalities (below 10), we find that using as little as 32 unique low discrepancy points performs similar or better than uniform sampling. In 2D, for which we have highly optimized low discrepancy samples available, we demonstrate that using these points yields the highest empirical performance and requires only 16 samples to improve over uniform sampling. Overall, we establish a clear relation between the $L_2$ discrepancy of the used point set and the empirical performance of the CMA-ES. - 7 September 2024
- Published by Springer Nature
- p. 268-283
Avoiding Redundant Restarts in Multimodal Global Optimization
- 7 September 2024
- Published by Springer Nature
- p. 268-283
Avoiding Redundant Restarts in Multimodal Global Optimization
- Top Cited Papers
- 3 September 2024
- Published by MIT Press in Evolutionary Computation
- Vol. 32 (3) , 205-210
IOHexperimenter: Benchmarking Platform for Iterative Optimization Heuristics
We present IOHexperimenter, the experimentation module of the IOHprofiler project. IOHexperimenter aims at providing an easy-to-use and customizable toolbox for benchmarking iterative optimization heuristics such as local search, evolutionary and genetic algorithms, and Bayesian optimization techniques. IOHexperimenter can be used as a stand-alone tool or as part of a benchmarking pipeline that uses other modules of the IOHprofiler environment. IOHexperimenter provides an efficient interface between optimization problems and their solvers while allowing for granular logging of the optimization process. Its logs are fully compatible with existing tools for interactive data analysis, which significantly speeds up the deployment of a benchmarking pipeline. The main components of IOHexperimenter are the environment to build customized problem suites and the various logging options that allow users to steer the granularity of the data records.
- 3 September 2024
- Published by MIT Press in Evolutionary Computation
- Vol. 32 (3) , 205-210
Top Cited PapersIOHexperimenter: Benchmarking Platform for Iterative Optimization Heuristics
We present IOHexperimenter, the experimentation module of the IOHprofiler project. IOHexperimenter aims at providing an easy-to-use and customizable toolbox for benchmarking iterative optimization heuristics such as local search, evolutionary and genetic algorithms, and Bayesian optimization techniques. IOHexperimenter can be used as a stand-alone tool or as part of a benchmarking pipeline that uses other modules of the IOHprofiler environment. IOHexperimenter provides an efficient interface between optimization problems and their solvers while allowing for granular logging of the optimization process. Its logs are fully compatible with existing tools for interactive data analysis, which significantly speeds up the deployment of a benchmarking pipeline. The main components of IOHexperimenter are the environment to build customized problem suites and the various logging options that allow users to steer the granularity of the data records.
- Open Access
- 14 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 791-799
Benchmarking and Analyzing Iterative Optimization Heuristics with IOHprofiler
- 14 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 791-799
Open AccessBenchmarking and Analyzing Iterative Optimization Heuristics with IOHprofiler
- Open Access
- 14 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 787-795
What Performance Indicators to Use for Self-Adaptation in Multi-Objective Evolutionary Algorithms
Parameter control has succeeded in accelerating the convergence process of evolutionary algorithms. While empirical and theoretical studies have shed light on the behavior of algorithms for single-objective optimization, little is known about how self-adaptation influences multi-objective evolutionary algorithms. In this work, we contribute (1) extensive experimental analysis of the Global Simple Evolutionary Multi-objective Algorithm (GSEMO) variants on classic problems, such as OneMinMax, LOTZ, COCZ, and (2) a novel version of GSEMO with self-adjusting mutation rates. To enable self-adaptation in GSEMO, we explore three techniques from single-objective optimization for self-adjusting mutation rates and use various performance metrics, such as hypervolume and inverted generational distance, to guide the adaptation. Our experiments show that adapting the mutation rate based on single-objective optimization and hypervolume can speed up the convergence of GSEMO. Moreover, we demonstrate that a GSEMO with self-adjusting mutation rates, which focuses on optimizing one of the objectives alternatively and adjusts the mutation rate for each solution individually, can outperform the GSEMO with static mutation rates across the tested problem. This work provides a comprehensive benchmarking study for MOEAs and complements existing theoretical runtime analysis. Our proposed algorithm addresses interesting issues for designing MOEAs for future practical applications.
- 14 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 787-795
Open AccessWhat Performance Indicators to Use for Self-Adaptation in Multi-Objective Evolutionary Algorithms
Parameter control has succeeded in accelerating the convergence process of evolutionary algorithms. While empirical and theoretical studies have shed light on the behavior of algorithms for single-objective optimization, little is known about how self-adaptation influences multi-objective evolutionary algorithms. In this work, we contribute (1) extensive experimental analysis of the Global Simple Evolutionary Multi-objective Algorithm (GSEMO) variants on classic problems, such as OneMinMax, LOTZ, COCZ, and (2) a novel version of GSEMO with self-adjusting mutation rates. To enable self-adaptation in GSEMO, we explore three techniques from single-objective optimization for self-adjusting mutation rates and use various performance metrics, such as hypervolume and inverted generational distance, to guide the adaptation. Our experiments show that adapting the mutation rate based on single-objective optimization and hypervolume can speed up the convergence of GSEMO. Moreover, we demonstrate that a GSEMO with self-adjusting mutation rates, which focuses on optimizing one of the objectives alternatively and adjusts the mutation rate for each solution individually, can outperform the GSEMO with static mutation rates across the tested problem. This work provides a comprehensive benchmarking study for MOEAs and complements existing theoretical runtime analysis. Our proposed algorithm addresses interesting issues for designing MOEAs for future practical applications.