Thomas Bäck
YOU?
Author Swipe
View article: From Performance to Understanding: A Vision for Explainable Automated Algorithm Design
From Performance to Understanding: A Vision for Explainable Automated Algorithm Design Open
Automated algorithm design is entering a new phase: Large Language Models can now generate full optimisation (meta)heuristics, explore vast design spaces and adapt through iterative feedback. Yet this rapid progress is largely performance-…
View article: Machine learning for hydrogen technologies: A comprehensive review of challenges, opportunities, and emerging trends
Machine learning for hydrogen technologies: A comprehensive review of challenges, opportunities, and emerging trends Open
View article: Benchmarking that Matters: Rethinking Benchmarking for Practical Impact
Benchmarking that Matters: Rethinking Benchmarking for Practical Impact Open
Benchmarking has driven scientific progress in Evolutionary Computation, yet current practices fall short of real-world needs. Widely used synthetic suites such as BBOB and CEC isolate algorithmic phenomena but poorly reflect the structure…
View article: From spikes to speech: NeuroVoc — A biologically plausible vocoder framework for auditory perception and cochlear implant simulation
From spikes to speech: NeuroVoc — A biologically plausible vocoder framework for auditory perception and cochlear implant simulation Open
We present NeuroVoc, a flexible model-agnostic vocoder framework that reconstructs acoustic waveforms from simulated neural activity patterns using an inverse Fourier transform. The system applies straightforward signal processing to neuro…
View article: REMAINING USEFUL LIFE IN COMPLEX MULTI-COMPONENT SYSTEMS: TAXONOMY, REVIEW, AND RESEARCH DIRECTIONS
REMAINING USEFUL LIFE IN COMPLEX MULTI-COMPONENT SYSTEMS: TAXONOMY, REVIEW, AND RESEARCH DIRECTIONS Open
View article: Multi-Step Reasoning with Large Language Models, a Survey
Multi-Step Reasoning with Large Language Models, a Survey Open
Large language models (LLMs) with billions of parameters exhibit in-context learning abilities, enabling few-shot learning on tasks that the model was not specifically trained for. Traditional models achieve breakthrough performance on lan…
View article: PATH: A discrete-sequence dataset for evaluating online unsupervised anomaly detection approaches for multivariate time series
PATH: A discrete-sequence dataset for evaluating online unsupervised anomaly detection approaches for multivariate time series Open
View article: DQS: A Low-Budget Query Strategy for Enhancing Unsupervised Data-driven Anomaly Detection Approaches
DQS: A Low-Budget Query Strategy for Enhancing Unsupervised Data-driven Anomaly Detection Approaches Open
Truly unsupervised approaches for time series anomaly detection are rare in the literature. Those that exist suffer from a poorly set threshold, which hampers detection performance, while others, despite claiming to be unsupervised, need t…
View article: BLADE: Benchmark suite for LLM-driven Automated Design and Evolution of iterative optimisation heuristics
BLADE: Benchmark suite for LLM-driven Automated Design and Evolution of iterative optimisation heuristics Open
View article: Abnormal Mutations: Evolution Strategies Don't Require Gaussianity
Abnormal Mutations: Evolution Strategies Don't Require Gaussianity Open
View article: Code Evolution Graphs: Understanding Large Language Model Driven Design of Algorithms
Code Evolution Graphs: Understanding Large Language Model Driven Design of Algorithms Open
View article: Evolving Hard Maximum Cut Instances for Quantum Approximate Optimization Algorithms
Evolving Hard Maximum Cut Instances for Quantum Approximate Optimization Algorithms Open
View article: Behaviour Space Analysis of LLM-driven Meta-heuristic Discovery
Behaviour Space Analysis of LLM-driven Meta-heuristic Discovery Open
We investigate the behaviour space of meta-heuristic optimisation algorithms automatically generated by Large Language Model driven algorithm discovery methods. Using the Large Language Evolutionary Algorithm (LLaMEA) framework with a GPT …
View article: Leveraging Lightweight Generators for Memory Efficient Continual Learning
Leveraging Lightweight Generators for Memory Efficient Continual Learning Open
Catastrophic forgetting can be trivially alleviated by keeping all data from previous tasks in memory. Therefore, minimizing the memory footprint while maximizing the amount of relevant information is crucial to the challenge of continual …
View article: Feasibility-Driven Trust Region Bayesian Optimization
Feasibility-Driven Trust Region Bayesian Optimization Open
Bayesian optimization is a powerful tool for solving real-world optimization tasks under tight evaluation budgets, making it well-suited for applications involving costly simulations or experiments. However, many of these tasks are also ch…
View article: LLaMEA-BO: A Large Language Model Evolutionary Algorithm for Automatically Generating Bayesian Optimization Algorithms
LLaMEA-BO: A Large Language Model Evolutionary Algorithm for Automatically Generating Bayesian Optimization Algorithms Open
Bayesian optimization (BO) is a powerful class of algorithms for optimizing expensive black-box functions, but designing effective BO algorithms remains a manual, expertise-driven task. Recent advancements in Large Language Models (LLMs) h…
View article: Evolutionary Computation and Large Language Models: A Survey of Methods, Synergies, and Applications
Evolutionary Computation and Large Language Models: A Survey of Methods, Synergies, and Applications Open
Integrating Large Language Models (LLMs) and Evolutionary Computation (EC) represents a promising avenue for advancing artificial intelligence by combining powerful natural language understanding with optimization and search capabilities. …
View article: Applying causality to environmental security in Iraq
Applying causality to environmental security in Iraq Open
View article: The travelling salesperson problem and the challenges of near-term quantum advantage
The travelling salesperson problem and the challenges of near-term quantum advantage Open
Over the last two decades, the travelling salesperson problem (TSP) has been cited as a benchmark problem to demonstrate the advantage of quantum computers over conventional computers. Its advantage is that it is a well-studied NP-hard opt…
View article: Study of the \(B_{4/2}\) Anomaly in the Yrast States of \(^{167}\)Os
Study of the \(B_{4/2}\) Anomaly in the Yrast States of \(^{167}\)Os Open
In recent years, several cases of nuclei presenting the so-called “\(B_{4/2}\) anomaly” have been observed in the neutron-deficient region close to \(Z=50\) and \(Z=82\). In the last region, the osmium isotopic chain is of particular inter…
View article: Utility-aware Social Network Anonymization using Genetic Algorithms
Utility-aware Social Network Anonymization using Genetic Algorithms Open
Social networks may contain privacy-sensitive information about individuals. The objective of the network anonymization problem is to alter a given social network dataset such that the number of anonymous nodes in the social graph is maxim…
View article: Surrogate-based automated hyperparameter optimization for expensive automotive crashworthiness optimization
Surrogate-based automated hyperparameter optimization for expensive automotive crashworthiness optimization Open
In the automotive industry, solving crashworthiness optimization problems efficiently is crucial to minimize time and cost investment on expensive function evaluations, e.g., using simulation runs. Nonetheless, automotive crashworthiness o…
View article: Diffusion Models for Tabular Data: Challenges, Current Progress, and Future Directions
Diffusion Models for Tabular Data: Challenges, Current Progress, and Future Directions Open
In recent years, generative models have achieved remarkable performance across diverse applications, including image generation, text synthesis, audio creation, video generation, and data augmentation. Diffusion models have emerged as supe…
View article: Explainable Benchmarking for Iterative Optimization Heuristics
Explainable Benchmarking for Iterative Optimization Heuristics Open
Benchmarking heuristic algorithms is vital to understand under which conditions and on what kind of problems certain algorithms perform well. In most current research into heuristic optimization algorithms, only a very limited number of sc…
View article: Abnormal Mutations: Evolution Strategies Don't Require Gaussianity
Abnormal Mutations: Evolution Strategies Don't Require Gaussianity Open
The mutation process in evolution strategies has been interlinked with the normal distribution since its inception. Many lines of reasoning have been given for this strong dependency, ranging from maximum entropy arguments to the need for …
View article: Evolving Hard Maximum Cut Instances for Quantum Approximate Optimization Algorithms
Evolving Hard Maximum Cut Instances for Quantum Approximate Optimization Algorithms Open
Variational quantum algorithms, such as the Recursive Quantum Approximate Optimization Algorithm (RQAOA), have become increasingly popular, offering promising avenues for employing Noisy Intermediate-Scale Quantum devices to address challe…
View article: Transfer Learning of Surrogate Models: Integrating Domain Warping and Affine Transformations
Transfer Learning of Surrogate Models: Integrating Domain Warping and Affine Transformations Open
Surrogate models provide efficient alternatives to computationally demanding real world processes but often require large datasets for effective training. A promising solution to this limitation is the transfer of pre-trained surrogate mod…
View article: Multi-Objective Deep-Learning-based Biomechanical Deformable Image Registration with MOREA
Multi-Objective Deep-Learning-based Biomechanical Deformable Image Registration with MOREA Open
When choosing a deformable image registration (DIR) approach for images with large deformations and content mismatch, the realism of found transformations often needs to be traded off against the required runtime. DIR approaches using deep…
View article: Transfer Learning of Surrogate Models via Domain Affine Transformation Across Synthetic and Real-World Benchmarks
Transfer Learning of Surrogate Models via Domain Affine Transformation Across Synthetic and Real-World Benchmarks Open
Surrogate models are frequently employed as efficient substitutes for the costly execution of real-world processes. However, constructing a high-quality surrogate model often demands extensive data acquisition. A solution to this issue is …
View article: Comparative Analysis of Indicators for Multi-objective Diversity Optimization
Comparative Analysis of Indicators for Multi-objective Diversity Optimization Open
Indicator-based (multi-objective) diversity optimization aims at finding a set of near (Pareto)optimal solutions that maximizes a diversity indicator, where diversity is typically interpreted as the number of essentially different solution…