Bogdan Gabrys

Alexander Scriven, David Jacob Kedziora, Katarzyna Musial et al.

With most technical fields, there exists a delay between fundamental academic research and practical industrial uptake. Whilst some sciences have robust and well-established processes for commercialisation, such as the pharmaceutical practice of regimented drug trials, other fields face transitory periods in which fundamental academic advancements diffuse gradually into the space of commerce and industry. For the still relatively young field of Automated/Autonomous Machine Learning (AutoML/AutonoML), that transitory period is under way, spurred on by a burgeoning interest from broader society. Yet, to date, little research has been undertaken to assess the current state of this dissemination and its uptake. Thus, this review makes two primary contributions to knowledge around this topic. Firstly, it provides the most up-to-date and comprehensive survey of existing AutoML tools, both open-source and commercial. Secondly, it motivates and outlines a framework for assessing whether an AutoML solution designed for real-world application is 'performant'; this framework extends beyond the limitations of typical academic criteria, considering a variety of stakeholder needs and the human-computer interactions required to service them. Thus, additionally supported by an extensive assessment and comparison of academic and commercial case-studies, this review evaluates mainstream engagement with AutoML in the early 2020s, identifying obstacles and opportunities for accelerating future uptake.

Thanh Tung Khuat, Bogdan Gabrys

Hyperbox-based machine learning algorithms are an important and popular branch of machine learning in the construction of classifiers using fuzzy sets and logic theory and neural network architectures. This type of learning is characterised by many strong points of modern predictors such as a high scalability, explainability, online adaptation, effective learning from a small amount of data, native ability to deal with missing data and accommodating new classes. Nevertheless, there is no comprehensive existing package for hyperbox-based machine learning which can serve as a benchmark for research and allow non-expert users to apply these algorithms easily. hyperbox-brain is an open-source Python library implementing the leading hyperbox-based machine learning algorithms. This library exposes a unified API which closely follows and is compatible with the renowned scikit-learn and numpy toolboxes. The library may be installed from Python Package Index (PyPI) and the conda package manager and is distributed under the GPL-3 license. The source code, documentation, detailed tutorials, and the full descriptions of the API are available at https://uts-caslab.github.io/hyperbox-brain.

Thanh Tung Khuat, David Jacob Kedziora, Bogdan Gabrys

As automated machine learning (AutoML) systems continue to progress in both sophistication and performance, it becomes important to understand the `how' and `why' of human-computer interaction (HCI) within these frameworks, both current and expected. Such a discussion is necessary for optimal system design, leveraging advanced data-processing capabilities to support decision-making involving humans, but it is also key to identifying the opportunities and risks presented by ever-increasing levels of machine autonomy. Within this context, we focus on the following questions: (i) How does HCI currently look like for state-of-the-art AutoML algorithms, especially during the stages of development, deployment, and maintenance? (ii) Do the expectations of HCI within AutoML frameworks vary for different types of users and stakeholders? (iii) How can HCI be managed so that AutoML solutions acquire human trust and broad acceptance? (iv) As AutoML systems become more autonomous and capable of learning from complex open-ended environments, will the fundamental nature of HCI evolve? To consider these questions, we project existing literature in HCI into the space of AutoML; this connection has, to date, largely been unexplored. In so doing, we review topics including user-interface design, human-bias mitigation, and trust in artificial intelligence (AI). Additionally, to rigorously gauge the future of HCI, we contemplate how AutoML may manifest in effectively open-ended environments. This discussion necessarily reviews projected developmental pathways for AutoML, such as the incorporation of reasoning, although the focus remains on how and why HCI may occur in such a framework rather than on any implementational details. Ultimately, this review serves to identify key research directions aimed at better facilitating the roles and modes of human interactions with both current and future AutoML systems.

Mingshan Jia, Bogdan Gabrys, Katarzyna Musial

The mining and exploitation of graph structural information have been the focal points in the study of complex networks. Traditional structural measures in Network Science focus on the analysis and modelling of complex networks from the perspective of network structure, such as the centrality measures, the clustering coefficient, and motifs and graphlets, and they have become basic tools for studying and understanding graphs. In comparison, graph neural networks, especially graph convolutional networks (GCNs), are particularly effective at integrating node features into graph structures via neighbourhood aggregation and message passing, and have been shown to significantly improve the performances in a variety of learning tasks. These two classes of methods are, however, typically treated separately with limited references to each other. In this work, aiming to establish relationships between them, we provide a network science perspective of GCNs. Our novel taxonomy classifies GCNs from three structural information angles, i.e., the layer-wise message aggregation scope, the message content, and the overall learning scope. Moreover, as a prerequisite for reviewing GCNs via a network science perspective, we also summarise traditional structural measures and propose a new taxonomy for them. Finally and most importantly, we draw connections between traditional structural approaches and graph convolutional networks, and discuss potential directions for future research.

Jiaqi Wen, Bogdan Gabrys, Katarzyna Musial

This study proposes an extendable modelling framework for Digital Twin-Oriented Complex Networked Systems (DT-CNSs) with a goal of generating networks that faithfully represent real systems. Modelling process focuses on (i) features of nodes and (ii) interaction rules for creating connections that are built based on individual node's preferences. We conduct experiments on simulation-based DT-CNSs that incorporate various features and rules about network growth and different transmissibilities related to an epidemic spread on these networks. We present a case study on disaster resilience of social networks given an epidemic outbreak by investigating the infection occurrence within specific time and social distance. The experimental results show how different levels of the structural and dynamics complexities, concerned with feature diversity and flexibility of interaction rules respectively, influence network growth and epidemic spread. The analysis revealed that, to achieve maximum disaster resilience, mitigation policies should be targeted at nodes with preferred features as they have higher infection risks and should be the focus of the epidemic control.

Thanh Tung Khuat, Robert Bassett, Ellen Otte et al.

Biopharmaceutical products, particularly monoclonal antibodies (mAbs), have gained prominence in the pharmaceutical market due to their high specificity and efficacy. As these products are projected to constitute a substantial portion of global pharmaceutical sales, the application of machine learning models in mAb development and manufacturing is gaining momentum. This paper addresses the critical need for uncertainty quantification in machine learning predictions, particularly in scenarios with limited training data. Leveraging ensemble learning and Monte Carlo simulations, our proposed method generates additional input samples to enhance the robustness of the model in small training datasets. We evaluate the efficacy of our approach through two case studies: predicting antibody concentrations in advance and real-time monitoring of glucose concentrations during bioreactor runs using Raman spectra data. Our findings demonstrate the effectiveness of the proposed method in estimating the uncertainty levels associated with process performance predictions and facilitating real-time decision-making in biopharmaceutical manufacturing. This contribution not only introduces a novel approach for uncertainty quantification but also provides insights into overcoming challenges posed by small training datasets in bioprocess development. The evaluation demonstrates the effectiveness of our method in addressing key challenges related to uncertainty estimation within upstream cell cultivation, illustrating its potential impact on enhancing process control and product quality in the dynamic field of biopharmaceuticals.

Thanh Tung Khuat, Robert Bassett, Ellen Otte et al.

While machine learning (ML) has made significant contributions to the biopharmaceutical field, its applications are still in the early stages in terms of providing direct support for quality-by-design based development and manufacturing of biopharmaceuticals, hindering the enormous potential for bioprocesses automation from their development to manufacturing. However, the adoption of ML-based models instead of conventional multivariate data analysis methods is significantly increasing due to the accumulation of large-scale production data. This trend is primarily driven by the real-time monitoring of process variables and quality attributes of biopharmaceutical products through the implementation of advanced process analytical technologies. Given the complexity and multidimensionality of a bioproduct design, bioprocess development, and product manufacturing data, ML-based approaches are increasingly being employed to achieve accurate, flexible, and high-performing predictive models to address the problems of analytics, monitoring, and control within the biopharma field. This paper aims to provide a comprehensive review of the current applications of ML solutions in a bioproduct design, monitoring, control, and optimisation of upstream, downstream, and product formulation processes. Finally, this paper thoroughly discusses the main challenges related to the bioprocesses themselves, process data, and the use of machine learning models in biopharmaceutical process development and manufacturing. Moreover, it offers further insights into the adoption of innovative machine learning methods and novel trends in the development of new digital biopharma solutions.

David Jacob Kedziora, Anna Musiał, Wojciech Rudno-Rudziński et al.

The use of single-photon sources (SPSs) is central to numerous systems and devices proposed amidst a modern surge in quantum technology. However, manufacturing schemes remain imperfect, and single-photon emission purity must often be experimentally verified via interferometry. Such a process is typically slow and costly, which has motivated growing research into whether SPS quality can be more rapidly inferred from incomplete emission statistics. Hence, this study is a sequel to previous work that demonstrated significant uncertainty in the standard method of quality estimation, i.e. the least-squares fitting of a physically motivated function, and asks: can machine learning (ML) do better? The study leverages eight datasets obtained from measurements involving an exemplary quantum emitter, i.e. a single InGaAs/GaAs epitaxial quantum dot; these eight contexts predominantly vary in the intensity of the exciting laser. Specifically, via a form of `transfer learning', five ML models, three linear and two ensemble-based, are trained on data from seven of the contexts and tested on the eighth. Validation metrics quickly reveal that even a linear regressor can outperform standard fitting when it is tested on the same contexts it was trained on, but the success of transfer learning is less assured, even though statistical analysis, made possible by data augmentation, suggests its superiority as an early estimator. Accordingly, the study concludes by discussing future strategies for grappling with the problem of SPS context dissimilarity, e.g. feature engineering and model adaptation.

David Jacob Kedziora, Anna Musiał, Wojciech Rudno-Rudziński et al.

Novel methods for rapidly estimating single-photon source (SPS) quality have been promoted in recent literature to address the expensive and time-consuming nature of experimental validation via intensity interferometry. However, the frequent lack of uncertainty discussions and reproducible details raises concerns about their reliability. This study investigates the use of data augmentation, a machine learning technique, to supplement experimental data with bootstrapped samples and quantify the uncertainty of such estimates. Eight datasets obtained from measurements involving a single InGaAs/GaAs epitaxial quantum dot serve as a proof-of-principle example. Analysis of one of the SPS quality metrics derived from efficient histogram fitting of the synthetic samples, i.e. the probability of multi-photon emission events, reveals significant uncertainty contributed by stochastic variability in the Poisson processes that describe detection rates. Ignoring this source of error risks severe overconfidence in both early quality estimates and claims for state-of-the-art SPS devices. Additionally, this study finds that standard least-squares fitting is comparable to using a Poisson likelihood, and expanding averages show some promise for early estimation. Also, reducing background counts improves fitting accuracy but does not address the Poisson-process variability. Ultimately, data augmentation demonstrates its value in supplementing physical experiments; its benefit here is to emphasise the need for a cautious assessment of SPS quality.

Jiaqi Wen, Bogdan Gabrys, Katarzyna Musial

Building models of Complex Networked Systems (CNS) that can accurately represent reality forms an important research area. To be able to reflect real world systems, the modelling needs to consider not only the intensity of interactions between the entities but also features of all the elements of the system. This study aims to improve the expressive power of node features in Digital Twin-Oriented Complex Networked Systems (DT-CNSs) with heterogeneous feature representation principles. This involves representing features with crisp feature values and fuzzy sets, each describing the objective and the subjective inductions of the nodes' features and feature differences. Our empirical analysis builds DT-CNSs to recreate realistic physical contact networks in different countries from real node feature distributions based on various representation principles and an optimised feature preference. We also investigate their respective disaster resilience to an epidemic outbreak starting from the most popular node. The results suggest that the increasing flexibility of feature representation with fuzzy sets improves the expressive power and enables more accurate modelling. In addition, the heterogeneous features influence the network structure and the speed of the epidemic outbreak, requiring various mitigation policies targeted at different people.

David Jacob Kedziora, Tien-Dung Nguyen, Katarzyna Musial et al.

The automated machine learning (AutoML) process can require searching through complex configuration spaces of not only machine learning (ML) components and their hyperparameters but also ways of composing them together, i.e. forming ML pipelines. Optimisation efficiency and the model accuracy attainable for a fixed time budget suffer if this pipeline configuration space is excessively large. A key research question is whether it is both possible and practical to preemptively avoid costly evaluations of poorly performing ML pipelines by leveraging their historical performance for various ML tasks, i.e. meta-knowledge. The previous experience comes in the form of classifier/regressor accuracy rankings derived from either (1) a substantial but non-exhaustive number of pipeline evaluations made during historical AutoML runs, i.e. 'opportunistic' meta-knowledge, or (2) comprehensive cross-validated evaluations of classifiers/regressors with default hyperparameters, i.e. 'systematic' meta-knowledge. Numerous experiments with the AutoWeka4MCPS package suggest that (1) opportunistic/systematic meta-knowledge can improve ML outcomes, typically in line with how relevant that meta-knowledge is, and (2) configuration-space culling is optimal when it is neither too conservative nor too radical. However, the utility and impact of meta-knowledge depend critically on numerous facets of its generation and exploitation, warranting extensive analysis; these are often overlooked/underappreciated within AutoML and meta-learning literature. In particular, we observe strong sensitivity to the `challenge' of a dataset, i.e. whether specificity in choosing a predictor leads to significantly better performance. Ultimately, identifying `difficult' datasets, thus defined, is crucial to both generating informative meta-knowledge bases and understanding optimal search-space reduction strategies.

Johnny Peng, Thanh Tung Khuat, Ellen Otte et al.

In cell culture bioprocessing, real-time batch process monitoring (BPM) refers to the continuous tracking and analysis of key process variables such as viable cell density, nutrient levels, metabolite concentrations, and product titer throughout the duration of a batch run. This enables early detection of deviations and supports timely control actions to ensure optimal cell growth and product quality. BPM plays a critical role in ensuring the quality and regulatory compliance of biopharmaceutical manufacturing processes. However, the development of accurate soft sensors for BPM is hindered by key challenges, including limited historical data, infrequent feedback, heterogeneous process conditions, and high-dimensional sensory inputs. This study presents a comprehensive benchmarking analysis of machine learning (ML) methods designed to address these challenges, with a focus on learning from historical data with limited volume and relevance in the context of bioprocess monitoring. We evaluate multiple ML approaches including feature dimensionality reduction, online learning, and just-in-time learning across three datasets, one in silico dataset and two real-world experimental datasets. Our findings highlight the importance of training strategies in handling limited data and feedback, with batch learning proving effective in homogeneous settings, while just-in-time learning and online learning demonstrate superior adaptability in cold-start scenarios. Additionally, we identify key meta-features, such as feed media composition and process control strategies, that significantly impact model transferability. The results also suggest that integrating Raman-based predictions with lagged offline measurements enhances monitoring accuracy, offering a promising direction for future bioprocess soft sensor development.

Jakub Binkowski, Denis Janiak, Albert Sawczyn et al.

Large Language Models (LLMs) have demonstrated remarkable performance across various tasks but remain prone to hallucinations. Detecting hallucinations is essential for safety-critical applications, and recent methods leverage attention map properties to this end, though their effectiveness remains limited. In this work, we investigate the spectral features of attention maps by interpreting them as adjacency matrices of graph structures. We propose the $\text{LapEigvals}$ method, which utilises the top-$k$ eigenvalues of the Laplacian matrix derived from the attention maps as an input to hallucination detection probes. Empirical evaluations demonstrate that our approach achieves state-of-the-art hallucination detection performance among attention-based methods. Extensive ablation studies further highlight the robustness and generalisation of $\text{LapEigvals}$, paving the way for future advancements in the hallucination detection domain.

Johnny Peng, Thanh Tung Khuat, Katarzyna Musial et al.

Data is crucial for machine learning (ML) applications, yet acquiring large datasets can be costly and time-consuming, especially in complex, resource-intensive fields like biopharmaceuticals. A key process in this industry is upstream bioprocessing, where living cells are cultivated and optimised to produce therapeutic proteins and biologics. The intricate nature of these processes, combined with high resource demands, often limits data collection, resulting in smaller datasets. This comprehensive review explores ML methods designed to address the challenges posed by small data and classifies them into a taxonomy to guide practical applications. Furthermore, each method in the taxonomy was thoroughly analysed, with a detailed discussion of its core concepts and an evaluation of its effectiveness in tackling small data challenges, as demonstrated by application results in the upstream bioprocessing and other related domains. By analysing how these methods tackle small data challenges from different perspectives, this review provides actionable insights, identifies current research gaps, and offers guidance for leveraging ML in data-constrained environments.

Denis Janiak, Jakub Binkowski, Albert Sawczyn et al.

Large language models (LLMs) have revolutionized natural language processing, yet their tendency to hallucinate poses serious challenges for reliable deployment. Despite numerous hallucination detection methods, their evaluations often rely on ROUGE, a metric based on lexical overlap that misaligns with human judgments. Through comprehensive human studies, we demonstrate that while ROUGE exhibits high recall, its extremely low precision leads to misleading performance estimates. In fact, several established detection methods show performance drops of up to 45.9\% when assessed using human-aligned metrics like LLM-as-Judge. Moreover, our analysis reveals that simple heuristics based on response length can rival complex detection techniques, exposing a fundamental flaw in current evaluation practices. We argue that adopting semantically aware and robust evaluation frameworks is essential to accurately gauge the true performance of hallucination detection methods, ultimately ensuring the trustworthiness of LLM outputs.

Albert Sawczyn, Jakub Binkowski, Denis Janiak et al.

Large Language Models (LLMs) frequently generate hallucinated content, posing significant challenges for applications where factuality is crucial. While existing hallucination detection methods typically operate at the sentence level or passage level, we propose FactSelfCheck, a novel black-box sampling-based method that enables fine-grained fact-level detection. Our approach represents text as knowledge graphs consisting of facts in the form of triples. Through analyzing factual consistency across multiple LLM responses, we compute fine-grained hallucination scores without requiring external resources or training data. Our evaluation demonstrates that FactSelfCheck performs competitively with leading sentence-level sampling-based methods while providing more detailed insights. Most notably, our fact-level approach significantly improves hallucination correction, achieving a 35.5% increase in factual content compared to the baseline, while sentence-level SelfCheckGPT yields only a 10.6% improvement. The granular nature of our detection enables more precise identification and correction of hallucinated content. Additionally, we contribute a new dataset for evaluating sampling-based methods - FavaMultiSamples.

Ali Nik-Khorasani, Thanh Tung Khuat, Bogdan Gabrys

This paper addresses the challenges of predicting bioprocess performance, particularly in monoclonal antibody (mAb) production, where conventional statistical methods often fall short due to time-series data's complexity and high dimensionality. We propose a novel Hyperbox Mixture Regression (HMR) model which employs hyperbox-based input space partitioning to enhance predictive accuracy while managing uncertainty inherent in bioprocess data. The HMR model is designed to dynamically generate hyperboxes for input samples in a single-pass process, thereby improving learning speed and reducing computational complexity. Our experimental study utilizes a dataset that contains 106 bioreactors. This study evaluates the model's performance in predicting critical quality attributes in monoclonal antibody manufacturing over a 15-day cultivation period. The results demonstrate that the HMR model outperforms comparable approximators in accuracy and learning speed and maintains interpretability and robustness under uncertain conditions. These findings underscore the potential of HMR as a powerful tool for enhancing predictive analytics in bioprocessing applications.

Thanh Tung Khuat, Johnny Peng, Robert Bassett et al.

This study explores the deployment of three machine learning (ML) approaches for real-time prediction of glucose, lactate, and ammonium concentrations in cell culture processes, using Raman spectroscopy as input features. The research addresses challenges associated with limited data availability and process variability, providing a comparative analysis of pretrained models, just-in-time learning (JITL), and online learning algorithms. Two industrial case studies are presented to evaluate the impact of varying bioprocess conditions on model performance. The findings highlight the specific conditions under which pretrained models demonstrate superior predictive accuracy and identify scenarios where JITL or online learning approaches are more effective for adaptive process monitoring. This study also highlights the critical importance of updating the deployed models/agents with the latest offline analytical measurements during bioreactor operations to maintain the model performance against the changes in cell growth behaviours and operating conditions throughout the bioreactor run. Additionally, the study confirms the usefulness of a simple mixture-of-experts framework in achieving enhanced accuracy and robustness for real-time predictions of metabolite concentrations based on Raman spectral data. These insights contribute to the development of robust strategies for the efficient deployment of ML models in dynamic and changing biomanufacturing environments.

Jiaqi Wen, Bogdan Gabrys, Katarzyna Musial

The Digital Twin Oriented Complex Networked System (DT-CNS) aims to build and extend a Complex Networked System (CNS) model with progressively increasing dynamics complexity towards an accurate reflection of reality -- a Digital Twin of reality. Our previous work proposed evolutionary DT-CNSs to model the long-term adaptive network changes in an epidemic outbreak. This study extends this framework by proposeing the temporal DT-CNS model, where reinforcement learning-driven nodes make decisions on temporal directed interactions in an epidemic outbreak. We consider cooperative nodes, as well as egocentric and ignorant "free-riders" in the cooperation. We describe this epidemic spreading process with the Susceptible-Infected-Recovered ($SIR$) model and investigate the impact of epidemic severity on the epidemic resilience for different types of nodes. Our experimental results show that (i) the full cooperation leads to a higher reward and lower infection number than a cooperation with egocentric or ignorant "free-riders"; (ii) an increasing number of "free-riders" in a cooperation leads to a smaller reward, while an increasing number of egocentric "free-riders" further escalate the infection numbers and (iii) higher infection rates and a slower recovery weakens networks' resilience to severe epidemic outbreaks. These findings also indicate that promoting cooperation and reducing "free-riders" can improve public health during epidemics.

Jiaqi Wen, Bogdan Gabrys, Katarzyna Musial

This paper aims to provide a comprehensive critical overview on how entities and their interactions in Complex Networked Systems (CNS) are modelled across disciplines as they approach their ultimate goal of creating a Digital Twin (DT) that perfectly matches the reality. We propose a new framework to conceptually compare diverse existing modelling paradigms from different perspectives and create unified assessment criteria to assess their respective capabilities of reaching such an ultimate goal. Using the proposed criteria, we also appraise how far the reviewed current state-of-the-art approaches are from the idealised DTs. We also identify and propose potential directions and ways of building a DT-orientated CNS based on the convergence and integration of CNS and DT utilising a variety of cross-disciplinary techniques.

Xuanyi Dong, David Jacob Kedziora, Katarzyna Musial et al.

Deep learning (DL) has proven to be a highly effective approach for developing models in diverse contexts, including visual perception, speech recognition, and machine translation. However, the end-to-end process for applying DL is not trivial. It requires grappling with problem formulation and context understanding, data engineering, model development, deployment, continuous monitoring and maintenance, and so on. Moreover, each of these steps typically relies heavily on humans, in terms of both knowledge and interactions, which impedes the further advancement and democratization of DL. Consequently, in response to these issues, a new field has emerged over the last few years: automated deep learning (AutoDL). This endeavor seeks to minimize the need for human involvement and is best known for its achievements in neural architecture search (NAS), a topic that has been the focus of several surveys. That stated, NAS is not the be-all and end-all of AutoDL. Accordingly, this review adopts an overarching perspective, examining research efforts into automation across the entirety of an archetypal DL workflow. In so doing, this work also proposes a comprehensive set of ten criteria by which to assess existing work in both individual publications and broader research areas. These criteria are: novelty, solution quality, efficiency, stability, interpretability, reproducibility, engineering quality, scalability, generalizability, and eco-friendliness. Thus, ultimately, this review provides an evaluative overview of AutoDL in the early 2020s, identifying where future opportunities for progress may exist.

Tien-Dung Nguyen, David Jacob Kedziora, Katarzyna Musial et al.

Machine learning (ML) pipeline composition and optimisation have been studied to seek multi-stage ML models, i.e. preprocessor-inclusive, that are both valid and well-performing. These processes typically require the design and traversal of complex configuration spaces consisting of not just individual ML components and their hyperparameters, but also higher-level pipeline structures that link these components together. Optimisation efficiency and resulting ML-model accuracy both suffer if this pipeline search space is unwieldy and excessively large; it becomes an appealing notion to avoid costly evaluations of poorly performing ML components ahead of time. Accordingly, this paper investigates whether, based on previous experience, a pool of available classifiers/regressors can be preemptively culled ahead of initiating a pipeline composition/optimisation process for a new ML problem, i.e. dataset. The previous experience comes in the form of classifier/regressor accuracy rankings derived, with loose assumptions, from a substantial but non-exhaustive number of pipeline evaluations; this meta-knowledge is considered 'opportunistic'. Numerous experiments with the AutoWeka4MCPS package, including ones leveraging similarities between datasets via the relative landmarking method, show that, despite its seeming unreliability, opportunistic meta-knowledge can improve ML outcomes. However, results also indicate that the culling of classifiers/regressors should not be too severe either. In effect, it is better to search through a 'top tier' of recommended predictors than to pin hopes onto one previously supreme performer.

David Jacob Kedziora, Katarzyna Musial, Bogdan Gabrys

Over the last decade, the long-running endeavour to automate high-level processes in machine learning (ML) has risen to mainstream prominence, stimulated by advances in optimisation techniques and their impact on selecting ML models/algorithms. Central to this drive is the appeal of engineering a computational system that both discovers and deploys high-performance solutions to arbitrary ML problems with minimal human interaction. Beyond this, an even loftier goal is the pursuit of autonomy, which describes the capability of the system to independently adjust an ML solution over a lifetime of changing contexts. However, these ambitions are unlikely to be achieved in a robust manner without the broader synthesis of various mechanisms and theoretical frameworks, which, at the present time, remain scattered across numerous research threads. Accordingly, this review seeks to motivate a more expansive perspective on what constitutes an automated/autonomous ML system, alongside consideration of how best to consolidate those elements. In doing so, we survey developments in the following research areas: hyperparameter optimisation, multi-component models, neural architecture search, automated feature engineering, meta-learning, multi-level ensembling, dynamic adaptation, multi-objective evaluation, resource constraints, flexible user involvement, and the principles of generalisation. We also develop a conceptual framework throughout the review, augmented by each topic, to illustrate one possible way of fusing high-level mechanisms into an autonomous ML system. Ultimately, we conclude that the notion of architectural integration deserves more discussion, without which the field of automated ML risks stifling both its technical advantages and general uptake.

Tien-Dung Nguyen, Bogdan Gabrys, Katarzyna Musial

Automated machine learning pipeline (ML) composition and optimisation aim at automating the process of finding the most promising ML pipelines within allocated resources (i.e., time, CPU and memory). Existing methods, such as Bayesian-based and genetic-based optimisation, which are implemented in Auto-Weka, Auto-sklearn and TPOT, evaluate pipelines by executing them. Therefore, the pipeline composition and optimisation of these methods frequently require a tremendous amount of time that prevents them from exploring complex pipelines to find better predictive models. To further explore this research challenge, we have conducted experiments showing that many of the generated pipelines are invalid in the first place, and attempting to execute them is a waste of time and resources. To address this issue, we propose a novel method to evaluate the validity of ML pipelines, without their execution, using a surrogate model (AVATAR). The AVATAR generates a knowledge base by automatically learning the capabilities and effects of ML algorithms on datasets' characteristics. This knowledge base is used for a simplified mapping from an original ML pipeline to a surrogate model which is a Petri net based pipeline. Instead of executing the original ML pipeline to evaluate its validity, the AVATAR evaluates its surrogate model constructed by capabilities and effects of the ML pipeline components and input/output simplified mappings. Evaluating this surrogate model is less resource-intensive than the execution of the original pipeline. As a result, the AVATAR enables the pipeline composition and optimisation methods to evaluate more pipelines by quickly rejecting invalid pipelines. We integrate the AVATAR into the sequential model-based algorithm configuration (SMAC). Our experiments show that when SMAC employs AVATAR, it finds better solutions than on its own.

Thanh Tung Khuat, Bogdan Gabrys

General fuzzy min-max neural network (GFMMNN) is one of the efficient neuro-fuzzy systems for data classification. However, one of the downsides of its original learning algorithms is the inability to handle and learn from the mixed-attribute data. While categorical features encoding methods can be used with the GFMMNN learning algorithms, they exhibit a lot of shortcomings. Other approaches proposed in the literature are not suitable for on-line learning as they require entire training data available in the learning phase. With the rapid change in the volume and velocity of streaming data in many application areas, it is increasingly required that the constructed models can learn and adapt to the continuous data changes in real-time without the need for their full retraining or access to the historical data. This paper proposes an extended online learning algorithm for the GFMMNN. The proposed method can handle the datasets with both continuous and categorical features. The extensive experiments confirmed superior and stable classification performance of the proposed approach in comparison to other relevant learning algorithms for the GFMM model.

Thanh Tung Khuat, Bogdan Gabrys

A general fuzzy min-max (GFMM) neural network is one of the efficient neuro-fuzzy systems for classification problems. However, a disadvantage of most of the current learning algorithms for GFMM is that they can handle effectively numerical valued features only. Therefore, this paper provides some potential approaches to adapting GFMM learning algorithms for classification problems with mixed-type or only categorical features as they are very common in practical applications and often carry very useful information. We will compare and assess three main methods of handling datasets with mixed features, including the use of encoding methods, the combination of the GFMM model with other classifiers, and employing the specific learning algorithms for both types of features. The experimental results showed that the target and James-Stein are appropriate categorical encoding methods for learning algorithms of GFMM models, while the combination of GFMM neural networks and decision trees is a flexible way to enhance the classification performance of GFMM models on datasets with the mixed features. The learning algorithms with the mixed-type feature abilities are potential approaches to deal with mixed-attribute data in a natural way, but they need further improvement to achieve a better classification accuracy. Based on the analysis, we also identify the strong and weak points of different methods and propose potential research directions.

Xuanyi Dong, Lu Liu, Katarzyna Musial et al.

Neural architecture search (NAS) has attracted a lot of attention and has been illustrated to bring tangible benefits in a large number of applications in the past few years. Architecture topology and architecture size have been regarded as two of the most important aspects for the performance of deep learning models and the community has spawned lots of searching algorithms for both aspects of the neural architectures. However, the performance gain from these searching algorithms is achieved under different search spaces and training setups. This makes the overall performance of the algorithms to some extent incomparable and the improvement from a sub-module of the searching model unclear. In this paper, we propose NATS-Bench, a unified benchmark on searching for both topology and size, for (almost) any up-to-date NAS algorithm. NATS-Bench includes the search space of 15,625 neural cell candidates for architecture topology and 32,768 for architecture size on three datasets. We analyze the validity of our benchmark in terms of various criteria and performance comparison of all candidates in the search space. We also show the versatility of NATS-Bench by benchmarking 13 recent state-of-the-art NAS algorithms on it. All logs and diagnostic information trained using the same setup for each candidate are provided. This facilitates a much larger community of researchers to focus on developing better NAS algorithms in a more comparable and computationally cost friendly environment. All codes are publicly available at: https://xuanyidong.com/assets/projects/NATS-Bench.

Abbas Raza Ali, Marcin Budka, Bogdan Gabrys

The exponential growth of volume, variety and velocity of data is raising the need for investigations of automated or semi-automated ways to extract useful patterns from the data. It requires deep expert knowledge and extensive computational resources to find the most appropriate mapping of learning methods for a given problem. It becomes a challenge in the presence of numerous configurations of learning algorithms on massive amounts of data. So there is a need for an intelligent recommendation engine that can advise what is the best learning algorithm for a dataset. The techniques that are commonly used by experts are based on a trial and error approach evaluating and comparing a number of possible solutions against each other, using their prior experience on a specific domain, etc. The trial and error approach combined with the expert's prior knowledge, though computationally and time expensive, have been often shown to work for stationary problems where the processing is usually performed off-line. However, this approach would not normally be feasible to apply to non-stationary problems where streams of data are continuously arriving. Furthermore, in a non-stationary environment, the manual analysis of data and testing of various methods whenever there is a change in the underlying data distribution would be very difficult or simply infeasible. In that scenario and within an on-line predictive system, there are several tasks where Meta-learning can be used to effectively facilitate best recommendations including 1) pre-processing steps, 2) learning algorithms or their combination, 3) adaptivity mechanisms and their parameters, 4) recurring concept extraction, and 5) concept drift detection.

Neil Vaughan, Bogdan Gabrys

This research proposes and evaluates scoring and assessment methods for Virtual Reality (VR) training simulators. VR simulators capture detailed n-dimensional human motion data which is useful for performance analysis. Custom made medical haptic VR training simulators were developed and used to record data from 271 trainees of multiple clinical experience levels. DTW Multivariate Prototyping (DTW-MP) is proposed. VR data was classified as Novice, Intermediate or Expert. Accuracy of algorithms applied for time-series classification were: dynamic time warping 1-nearest neighbor (DTW-1NN) 60%, nearest centroid SoftDTW classification 77.5%, Deep Learning: ResNet 85%, FCN 75%, CNN 72.5% and MCDCNN 28.5%. Expert VR data recordings can be used for guidance of novices. Assessment feedback can help trainees to improve skills and consistency. Motion analysis can identify different techniques used by individuals. Mistakes can be detected dynamically in real-time, raising alarms to prevent injuries.

Xuanyi Dong, Mingxing Tan, Adams Wei Yu et al.

Efficient hyperparameter or architecture search methods have shown remarkable results, but each of them is only applicable to searching for either hyperparameters (HPs) or architectures. In this work, we propose a unified pipeline, AutoHAS, to efficiently search for both architectures and hyperparameters. AutoHAS learns to alternately update the shared network weights and a reinforcement learning (RL) controller, which learns the probability distribution for the architecture candidates and HP candidates. A temporary weight is introduced to store the updated weight from the selected HPs (by the controller), and a validation accuracy based on this temporary weight serves as a reward to update the controller. In experiments, we show AutoHAS is efficient and generalizable to different search spaces, baselines and datasets. In particular, AutoHAS can improve the accuracy over popular network architectures, such as ResNet and EfficientNet, on CIFAR-10/100, ImageNet, and four more other datasets.

Hongxu Chen, Hongzhi Yin, Xiangguo Sun et al.

Cross-platform account matching plays a significant role in social network analytics, and is beneficial for a wide range of applications. However, existing methods either heavily rely on high-quality user generated content (including user profiles) or suffer from data insufficiency problem if only focusing on network topology, which brings researchers into an insoluble dilemma of model selection. In this paper, to address this problem, we propose a novel framework that considers multi-level graph convolutions on both local network structure and hypergraph structure in a unified manner. The proposed method overcomes data insufficiency problem of existing work and does not necessarily rely on user demographic information. Moreover, to adapt the proposed method to be capable of handling large-scale social networks, we propose a two-phase space reconciliation mechanism to align the embedding spaces in both network partitioning based parallel training and account matching across different social networks. Extensive experiments have been conducted on two large-scale real-life social networks. The experimental results demonstrate that the proposed method outperforms the state-of-the-art models with a big margin.

Thanh Tung Khuat, Bogdan Gabrys

This paper proposes a simple yet powerful ensemble classifier, called Random Hyperboxes, constructed from individual hyperbox-based classifiers trained on the random subsets of sample and feature spaces of the training set. We also show a generalization error bound of the proposed classifier based on the strength of the individual hyperbox-based classifiers as well as the correlation among them. The effectiveness of the proposed classifier is analyzed using a carefully selected illustrative example and compared empirically with other popular single and ensemble classifiers via 20 datasets using statistical testing methods. The experimental results confirmed that our proposed method outperformed other fuzzy min-max neural networks, popular learning algorithms, and is competitive with other ensemble methods. Finally, we identify the existing issues related to the generalization error bounds of the real datasets and inform the potential research directions.

Joakim Skarding, Bogdan Gabrys, Katarzyna Musial

Dynamic networks are used in a wide range of fields, including social network analysis, recommender systems, and epidemiology. Representing complex networks as structures changing over time allow network models to leverage not only structural but also temporal patterns. However, as dynamic network literature stems from diverse fields and makes use of inconsistent terminology, it is challenging to navigate. Meanwhile, graph neural networks (GNNs) have gained a lot of attention in recent years for their ability to perform well on a range of network science tasks, such as link prediction and node classification. Despite the popularity of graph neural networks and the proven benefits of dynamic network models, there has been little focus on graph neural networks for dynamic networks. To address the challenges resulting from the fact that this research crosses diverse fields as well as to survey dynamic graph neural networks, this work is split into two main parts. First, to address the ambiguity of the dynamic network terminology we establish a foundation of dynamic networks with consistent, detailed terminology and notation. Second, we present a comprehensive survey of dynamic graph neural network models using the proposed terminology

Thanh Tung Khuat, Bogdan Gabrys

This paper proposes a method to accelerate the training process of a general fuzzy min-max neural network. The purpose is to reduce the unsuitable hyperboxes selected as the potential candidates of the expansion step of existing hyperboxes to cover a new input pattern in the online learning algorithms or candidates of the hyperbox aggregation process in the agglomerative learning algorithms. Our proposed approach is based on the mathematical formulas to form a branch-and-bound solution aiming to remove the hyperboxes which are certain not to satisfy expansion or aggregation conditions, and in turn, decreasing the training time of learning algorithms. The efficiency of the proposed method is assessed over a number of widely used data sets. The experimental results indicated the significant decrease in training time of the proposed approach for both online and agglomerative learning algorithms. Notably, the training time of the online learning algorithms is reduced from 1.2 to 12 times when using the proposed method, while the agglomerative learning algorithms are accelerated from 7 to 37 times on average.

Tien-Dung Nguyen, Tomasz Maszczyk, Katarzyna Musial et al.

The evaluation of machine learning (ML) pipelines is essential during automatic ML pipeline composition and optimisation. The previous methods such as Bayesian-based and genetic-based optimisation, which are implemented in Auto-Weka, Auto-sklearn and TPOT, evaluate pipelines by executing them. Therefore, the pipeline composition and optimisation of these methods requires a tremendous amount of time that prevents them from exploring complex pipelines to find better predictive models. To further explore this research challenge, we have conducted experiments showing that many of the generated pipelines are invalid, and it is unnecessary to execute them to find out whether they are good pipelines. To address this issue, we propose a novel method to evaluate the validity of ML pipelines using a surrogate model (AVATAR). The AVATAR enables to accelerate automatic ML pipeline composition and optimisation by quickly ignoring invalid pipelines. Our experiments show that the AVATAR is more efficient in evaluating complex pipelines in comparison with the traditional evaluation approaches requiring their execution.

Thanh Tung Khuat, Fang Chen, Bogdan Gabrys

This paper proposes an improved version of the current online learning algorithm for a general fuzzy min-max neural network (GFMM) to tackle existing issues concerning expansion and contraction steps as well as the way of dealing with unseen data located on decision boundaries. These drawbacks lower its classification performance, so an improved algorithm is proposed in this study to address the above limitations. The proposed approach does not use the contraction process for overlapping hyperboxes, which is more likely to increase the error rate as shown in the literature. The empirical results indicated the improvement in the classification accuracy and stability of the proposed method compared to the original version and other fuzzy min-max classifiers. In order to reduce the sensitivity to the training samples presentation order of this new on-line learning algorithm, a simple ensemble method is also proposed.

Thanh Tung Khuat, Bogdan Gabrys

General fuzzy min-max (GFMM) neural network is a generalization of fuzzy neural networks formed by hyperbox fuzzy sets for classification and clustering problems. Two principle algorithms are deployed to train this type of neural network, i.e., incremental learning and agglomerative learning. This paper presents a comprehensive empirical study of performance influencing factors, advantages, and drawbacks of the general fuzzy min-max neural network on pattern classification problems. The subjects of this study include (1) the impact of maximum hyperbox size, (2) the influence of the similarity threshold and measures on the agglomerative learning algorithm, (3) the effect of data presentation order, (4) comparative performance evaluation of the GFMM with other types of fuzzy min-max neural networks and prevalent machine learning algorithms. The experimental results on benchmark datasets widely used in machine learning showed overall strong and weak points of the GFMM classifier. These outcomes also informed potential research directions for this class of machine learning algorithms in the future.

Thanh Tung Khuat, Fang Chen, Bogdan Gabrys

Motivated by the practical demands for simplification of data towards being consistent with human thinking and problem solving as well as tolerance of uncertainty, information granules are becoming important entities in data processing at different levels of data abstraction. This paper proposes a method to construct classifiers from multi-resolution hierarchical granular representations (MRHGRC) using hyperbox fuzzy sets. The proposed approach forms a series of granular inferences hierarchically through many levels of abstraction. An attractive characteristic of our classifier is that it can maintain relatively high accuracy at a low degree of granularity based on reusing the knowledge learned from lower levels of abstraction. In addition, our approach can reduce the data size significantly as well as handling the uncertainty and incompleteness associated with data in real-world applications. The construction process of the classifier consists of two phases. The first phase is to formulate the model at the greatest level of granularity, while the later stage aims to reduce the complexity of the constructed model and deduce it from data at higher abstraction levels. Experimental outcomes conducted comprehensively on both synthetic and real datasets indicated the efficiency of our method in terms of training time and predictive performance in comparison to other types of fuzzy min-max neural networks and common machine learning algorithms.

Thanh Tung Khuat, Dymitr Ruta, Bogdan Gabrys

With the rapid development of digital information, the data volume generated by humans and machines is growing exponentially. Along with this trend, machine learning algorithms have been formed and evolved continuously to discover new information and knowledge from different data sources. Learning algorithms using hyperboxes as fundamental representational and building blocks are a branch of machine learning methods. These algorithms have enormous potential for high scalability and online adaptation of predictors built using hyperbox data representations to the dynamically changing environments and streaming data. This paper aims to give a comprehensive survey of literature on hyperbox-based machine learning models. In general, according to the architecture and characteristic features of the resulting models, the existing hyperbox-based learning algorithms may be grouped into three major categories: fuzzy min-max neural networks, hyperbox-based hybrid models, and other algorithms based on hyperbox representations. Within each of these groups, this paper shows a brief description of the structure of models, associated learning algorithms, and an analysis of their advantages and drawbacks. Main applications of these hyperbox-based models to the real-world problems are also described in this paper. Finally, we discuss some open problems and identify potential future research directions in this field.

Rashid Bakirov, Bogdan Gabrys, Damien Fay

Automation of machine learning model development is increasingly becoming an established research area. While automated model selection and automated data pre-processing have been studied in depth, there is, however, a gap concerning automated model adaptation strategies when multiple strategies are available. Manually developing an adaptation strategy can be time consuming and costly. In this paper we address this issue by proposing the use of flexible adaptive mechanism deployment for automated development of adaptation strategies. Experimental results after using the proposed strategies with five adaptive algorithms on 36 datasets confirm their viability. These strategies achieve better or comparable performance to the custom adaptation strategies and the repeated deployment of any single adaptive mechanism.

Manuel Martin Salvador, Marcin Budka, Bogdan Gabrys

Composition and parameterization of multicomponent predictive systems (MCPSs) consisting of chains of data transformation steps are a challenging task. Auto-WEKA is a tool to automate the combined algorithm selection and hyperparameter (CASH) optimization problem. In this paper, we extend the CASH problem and Auto-WEKA to support the MCPS, including preprocessing steps for both classification and regression tasks. We define the optimization problem in which the search space consists of suitably parameterized Petri nets forming the sought MCPS solutions. In the experimental analysis, we focus on examining the impact of considerably extending the search space (from approximately 22,000 to 812 billion possible combinations of methods and categorical hyperparameters). In a range of extensive experiments, three different optimization strategies are used to automatically compose MCPSs for 21 publicly available data sets. The diversity of the composed MCPSs found is an indication that fully and automatically exploiting different combinations of data cleaning and preprocessing techniques is possible and highly beneficial for different predictive models. We also present the results on seven data sets from real chemical production processes. Our findings can have a major impact on the development of high-quality predictive models as well as their maintenance and scalability aspects needed in modern applications and deployment scenarios.