Defining the docking search space is a critical yet often overlooked step in molecular docking, especially for receptors that lack clear or well-structured binding pockets. We address this challenge by formulating grid box placement as a global, expensive black-box optimization problem and introduce Evolutionary Driven Bayesian Optimization (EA-BO), a surrogate-based framework designed for efficient exploration under strict evaluation budgets. EA-BO integrates Gaussian Process models with a Matérn 5/2 kernel, LBFGS-B hyperparameter tuning, and CMA-ES-driven acquisition maximization to balance exploration and exploitation in a computationally demanding setting. We evaluate EA-BO on the interleukin-6 receptor, where manual grid selection and standard heuristics frequently fail. Across a panel of resveratrol-like ligands selected through ECFP4-based similarity screening, EA-BO consistently identifies interaction hotspots and converges faster than Optuna, Gaussian Process Bayesian Optimization, and Scikit-Optimize, while also outperforming grid centers reported in previous IL-6R docking studies. As a second contribution, we leverage the optimized docking region to rank and select the best-performing ligand from the resveratrol analogue set based on binding affinity with the receptor, demonstrating how automated grid optimization directly facilitates ligand prioritization. These results demonstrate that EA-BO provides data-efficient strategy for locating promising docking regions when computational cost limit traditional approaches.

The optimal allocation of crops to different parcels of land is a problem of paramount practical importance, not only to improve food and feed production, but also to address the challenges posed by climate change. However, this optimization problem is inherently complex due to the large number of agricultural sites available which generates a vast search space that renders traditional optimization techniques impractical. Moreover, as maximizing average production may generate solutions characterized by high year-by-year instability and lead to large and unrealistic cultivated areas, it is necessary to optimize crop allocation considering several objectives at the same time. In order to tackle this complex optimization problem, we propose a multi-objective approach, simultaneously maximizing the average production, minimizing the year-on-year production variance, and minimizing the total cultivated surface. The approach relies on an established multi-objective evolutionary algorithm, and employs a machine learning model able to predict crop production from weather and irrigation conditions, trained on historical data, making it possible to tackle allocation problems of large size. The proposed approach is compared to a quadratic programming algorithm tailored to the target problem. A case study focusing on the allocation of soybean crops in the European continent for the years 2000–2023 shows that the proposed methodology is able to identify informative tradeoffs between the three conflicting objectives considered, and identify realistic and meaningful crop allocations for supporting stakeholders’ decisions.

The optimal allocation of crops to different parcels of land is a problem of paramount practical importance, not only to improve production, but also to address the challenges posed by climate change. However, this optimization problem is inherently complex, characterized by a vast search space that renders traditional optimization techniques impractical without oversimplified assumptions. Compounding this challenge, climate change introduces conflicting objectives, as solutions aiming to just maximize total yield may be more susceptible to extreme weather events, and thus obtain more unpredictable year-by-year outcomes. In order to tackle this complex optimization problem, we propose a multi-objective approach, simultaneously maximizing the overall yield, minimizing the year-on-year yield variance, and minimizing the total cultivated surface. The approach exploits an established multi-objective evolutionary algorithm, and employs a machine learning model able to predict yield from weather and soil conditions, trained on historical data, making it possible to tackle allocation problems of large size. An experimental evaluation focusing on the allocation of soybean crops in the European continent for the years 2000-2023 shows that the proposed methodology is able to identify different trade-offs between the conflicting objectives, that an expert analysis later reveals to be realistic and meaningful for driving stakeholder decisions.

Inflammatory Bowel Disease (IBD), that includes Crohn's disease (CD) and Ulcerative Colitis (UC), is a global health concern due to the increasing number of cases. Diagnosing IBD is a challenging task due to a considerable number of clinical factors. Delayed or inaccurate IBD diagnosis can worsen the disease and complicate achieving remission, therefore, early diagnosis and prompt treatment are crucial. In this study, we adapted a methodology to analyze 16s rRNA (18,758 features) to analyze mRNA (54,675 features) that consists of three phases: 1) preprocessing, 2) feature selection, and 3) testing. We applied this methodology for analyzing mRNA datasets from the Gene Expression Omnibus (GEO) repository, aiming to discover possible biomarkers for IBD diagnosis. We experimented with three datasets, using one dataset for feature (gene) selection and we tested the results in the other two. We compared results with those obtained from other feature selection methods, such as the F-score-based K-Best and random selection. The Area Under the Curve (AUC) was used to measure the diagnostic accuracy and as a metric to compare results between the methodology and other feature selection methods. The Matthews Correlation Coefficient (MCC) was used as an additional metric to evaluate the performance of the methodology and for comparison with other feature selection methods.

Interactive Evolutionary Systems (IES) can generate several designs based on a handful of input parameters. Never-theless, the choice of the parameters is an open problem and it is limited to a few evaluations as they require human input. As a solution, Bayesian Optimization (BO) can be used to tune IES parameters. BO is a statistical method that efficiently models and optimizes expensive black-box derivative-free functions in few evaluations. In the context of creative IES, such as image generators, it can be used in conjunction with user preferences to optimize a complex-structured input space, such as variations of artistic images with uniqueness and creativity that follow the original concept and the artistic intention. Therefore, for this objective, we propose an implementation of BOIES with a metric based on user preferences that interactively evaluates a batch of images to evolve a set of parameters in Stable Diffusion to create variations with a given human-made artwork. Our results proved better than baseline, and against generated images using Neural Style Transfer (NST). The resulting images were consistent in terms of uniqueness, quality, and following a given concept.

Federated Learning (FL), a privacy-aware approach in distributed deep learning environments, enables many clients to collaboratively train a model without sharing sensitive data, thereby reducing privacy risks. However, enabling human trust and control over FL systems requires understanding the evolving behaviour of clients, whether beneficial or detrimental for the training, which still represents a key challenge in the current literature. To address this challenge, we introduce Federated Behavioural Planes (FBPs), a novel method to analyse, visualise, and explain the dynamics of FL systems, showing how clients behave under two different lenses: predictive performance (error behavioural space) and decision-making processes (counterfactual behavioural space). Our experiments demonstrate that FBPs provide informative trajectories describing the evolving states of clients and their contributions to the global model, thereby enabling the identification of clusters of clients with similar behaviours. Leveraging the patterns identified by FBPs, we propose a robust aggregation technique named Federated Behavioural Shields to detect malicious or noisy client models, thereby enhancing security and surpassing the efficacy of existing state-of-the-art FL defense mechanisms. Our code is publicly available on GitHub.

BACKGROUND: In recent years, human microbiome studies have received increasing attention as this field is considered a potential source for clinical applications. With the advancements in omics technologies and AI, research focused on the discovery for potential biomarkers in the human microbiome using machine learning tools has produced positive outcomes. Despite the promising results, several issues can still be found in these studies such as datasets with small number of samples, inconsistent results, lack of uniform processing and methodologies, and other additional factors lead to lack of reproducibility in biomedical research. In this work, we propose a methodology that combines the DADA2 pipeline for 16s rRNA sequences processing and the Recursive Ensemble Feature Selection (REFS) in multiple datasets to increase reproducibility and obtain robust and reliable results in biomedical research. RESULTS: Three experiments were performed analyzing microbiome data from patients/cases in Inflammatory Bowel Disease (IBD), Autism Spectrum Disorder (ASD), and Type 2 Diabetes (T2D). In each experiment, we found a biomarker signature in one dataset and applied to 2 other as further validation. The effectiveness of the proposed methodology was compared with other feature selection methods such as K-Best with F-score and random selection as a base line. The Area Under the Curve (AUC) was employed as a measure of diagnostic accuracy and used as a metric for comparing the results of the proposed methodology with other feature selection methods. Additionally, we use the Matthews Correlation Coefficient (MCC) as a metric to evaluate the performance of the methodology as well as for comparison with other feature selection methods. CONCLUSIONS: We developed a methodology for reproducible biomarker discovery for 16s rRNA microbiome sequence analysis, addressing the issues related with data dimensionality, inconsistent results and validation across independent datasets. The findings from the three experiments, across 9 different datasets, show that the proposed methodology achieved higher accuracy compared to other feature selection methods. This methodology is a first approach to increase reproducibility, to provide robust and reliable results.

The mathematical theory of viability, developed to formalize problems related to natural and social phenomena, investigates the evolution of dynamical systems under constraints. A main objective of this theory is to design control laws to keep systems inside viable domains. Control laws are traditionally defined as rules, based on the current position in the state space with respect to the boundaries of the viability kernel. However, finding these boundaries is a computationally expensive procedure, feasible only for trivial systems. We propose an approach based on Genetic Programming (GP) to discover control laws for viability problems in analytic form. Such laws could keep a system viable without the need of computing its viability kernel, facilitate communication with stakeholders, and improve explainability. A candidate set of control rules is encoded as GP trees describing equations. Evaluation is noisy, due to stochastic sampling: initial conditions are randomly drawn from the state space of the problem, and for each, a system of differential equations describing the system is solved, creating a trajectory. Candidate control laws are rewarded for keeping viable as many trajectories as possible, for as long as possible. The proposed approach is evaluated on established benchmarks for viability and delivers promising results.

Deep learning methods are highly accurate, yet their opaque decision process prevents them from earning full human trust. Concept-based models aim to address this issue by learning tasks based on a set of human-understandable concepts. However, state-of-the-art concept-based models rely on high-dimensional concept embedding representations which lack a clear semantic meaning, thus questioning the interpretability of their decision process. To overcome this limitation, we propose the Deep Concept Reasoner (DCR), the first interpretable concept-based model that builds upon concept embeddings. In DCR, neural networks do not make task predictions directly, but they build syntactic rule structures using concept embeddings. DCR then executes these rules on meaningful concept truth degrees to provide a final interpretable and semantically-consistent prediction in a differentiable manner. Our experiments show that DCR: (i) improves up to +25% w.r.t. state-of-the-art interpretable concept-based models on challenging benchmarks (ii) discovers meaningful logic rules matching known ground truths even in the absence of concept supervision during training, and (iii), facilitates the generation of counterfactual examples providing the learnt rules as guidance.

The backlash is a lost motion in a mechanism created by gaps between its parts. It causes vibrations that increase over time and negatively affect accuracy and performance. The quickest and most precise way to measure the backlash is to use specific sensors, that have to be added to the standard equipment of the robot. However, this solution is little used in practice because raises the manufacturing costs. An alternative solution can be to exploit a virtual sensor, i.e., the information about phenomena that are not directly measured is reconstructed by signals from sensors used for other measurements.This work evaluates the use of bio-inspired swarm algorithms as the processing core of a virtual sensor for the backlash of a robotic joint. Swarm-based approaches, with their relatively modest occupation of memory and low computational load, could be ideal candidates to solve the problem. In this paper, we exploit four state-of-the-art swarm-based optimization algorithms: the Dragonfly Algorithm, the Ant Lion Optimizer, the Grasshopper Optimization Algorithm, and the Grey Wolf Optimizer. The four candidate algorithms are compared on 20 different datasets covering a range of backlash values that reflect an industrial case scenario. Numerical results indicate that, unfortunately, none of the algorithms considered provides satisfactory solutions for the problem analyzed. Therefore, even if promising, these algorithms cannot represent the final choice for the problem of interest.

Primer sets are short DNA sequences of 18-22 base pairs, that can be used to verify the presence of a virus, and designed to attach to a specific part of a viral DNA. Designing a primer set requires choosing a region of DNA, avoiding the possibility of hybridization to a similar sequence, as well as considering its GC content and Tm (melting temperature). Coronaviruses, such as SARS-CoV-2, have a considerably large genome (around 30 thousand nucleotides) when compared to other viruses. With the rapid rise and spread of SARS-CoV-2 variants, it has become a priority to breach our lack of specific primers available for diagnosis of this new variants. Here, we propose an evolutionary-based approach to primer design, able to rapidly deliver a high-quality primer set for a target sequence of the virus variant. Starting from viral sequences collected from open repositories, the proposed approach is proven able to uncover a specific primer set for the B.1.1.7 SARS-CoV-2 variant. Only recently identified, B.1.1.7 is already considered potentially dangerous, as it presents a considerably higher transmissibility when compared to other variants.

BACKGROUND: MicroRNAs (miRNAs) are noncoding RNA molecules heavily involved in human tumors, in which few of them circulating the human body. Finding a tumor-associated signature of miRNA, that is, the minimum miRNA entities to be measured for discriminating both different types of cancer and normal tissues, is of utmost importance. Feature selection techniques applied in machine learning can help however they often provide naive or biased results. RESULTS: An ensemble feature selection strategy for miRNA signatures is proposed. miRNAs are chosen based on consensus on feature relevance from high-accuracy classifiers of different typologies. This methodology aims to identify signatures that are considerably more robust and reliable when used in clinically relevant prediction tasks. Using the proposed method, a 100-miRNA signature is identified in a dataset of 8023 samples, extracted from TCGA. When running eight-state-of-the-art classifiers along with the 100-miRNA signature against the original 1046 features, it could be detected that global accuracy differs only by 1.4%. Importantly, this 100-miRNA signature is sufficient to distinguish between tumor and normal tissues. The approach is then compared against other feature selection methods, such as UFS, RFE, EN, LASSO, Genetic Algorithms, and EFS-CLA. The proposed approach provides better accuracy when tested on a 10-fold cross-validation with different classifiers and it is applied to several GEO datasets across different platforms with some classifiers showing more than 90% classification accuracy, which proves its cross-platform applicability. CONCLUSIONS: The 100-miRNA signature is sufficiently stable to provide almost the same classification accuracy as the complete TCGA dataset, and it is further validated on several GEO datasets, across different types of cancer and platforms. Furthermore, a bibliographic analysis confirms that 77 out of the 100 miRNAs in the signature appear in lists of circulating miRNAs used in cancer studies, in stem-loop or mature-sequence form. The remaining 23 miRNAs offer potentially promising avenues for future research.

The iterated prisoner's dilemma is a famous model of cooperation and conflict in game theory. Its origin can be traced back to the Cold War, and countless strategies for playing it have been proposed so far, either designed by hand or automatically generated by computers. In the 2000s, scholars started focusing on adaptive players, that is, able to classify their opponent's behavior and adopt an effective counter-strategy. The player presented in this paper, pushes such idea even further: it builds a model of the current adversary from scratch, without relying on any pre-defined archetypes, and tweaks it as the game develops using an evolutionary algorithm; at the same time, it exploits the model to lead the game into the most favorable continuation. Models are compact nondeterministic finite state machines; they are extremely efficient in predicting opponents' replies, without being completely correct by necessity. Experimental results show that such a player is able to win several one-to-one games against strong opponents taken from the literature, and that it consistently prevails in round-robin tournaments of different sizes.

Evolutionary algorithms greatly benefit from an optimal application of the different genetic operators during the optimization process: thus, it is not surprising that several research lines in literature deal with the self-adapting of activation probabilities for operators. The current state of the art revolves around the use of the Multi-Armed Bandit (MAB) and Dynamic Multi-Armed bandit (D-MAB) paradigms, that modify the selection mechanism based on the rewards of the different operators. Such methodologies, however, update the probabilities after each operator's application, creating possible issues with positive feedbacks and impairing parallel evaluations, one of the strongest advantages of evolutionary computation in an industrial perspective. Moreover, D-MAB techniques often rely upon measurements of population diversity, that might not be applicable to all real-world scenarios. In this paper, we propose a generalization of the D-MAB approach, paired with a simple mechanism for operator management, that aims at removing several limitations of other D-MAB strategies, allowing for parallel evaluations and self-adaptive parameter tuning. Experimental results show that the approach is particularly effective with frameworks containing many different operators, even when some of them are ill-suited for the problem at hand, or are sporadically failing, as it commonly happens in the real world.

The iterated prisoner's dilemma is a widely known model in game theory, fundamental to many theories of cooperation and trust among self-interested beings. There are many works in literature about developing efficient strategies for this problem, both inside and outside the machine learning community. This paper shift the focus from finding a “good strategy” in absolute terms, to dynamically adapting and optimizing the strategy against the current opponent. Turan evolves competitive non-deterministic models of the current opponent, and exploit them to predict its moves and maximize the payoff as the game develops. Experimental results show that the proposed approach is able to obtain good performances against different kind of opponent, whether their strategies can or cannot be implemented as finite state machines.

Wireless sensor network (WSN) routing protocols, e.g., the Collection Tree Protocol (CTP), are designed to adapt in an ad-hoc fashion to the quality of the environment. WSNs thus have high internal dynamics and complex global behavior. Classical techniques for performance evaluation (such as testing or verification) fail to uncover the cases of extreme behavior which are most interesting to designers. We contribute a practical framework for performance evaluation of WSN protocols. The framework is based on multi-objective optimization, coupled with protocol simulation and evaluation of performance factors. For evaluation, we consider the two crucial functional and non-functional performance factors of a WSN, respectively: the ratio of data delivery from the network (DDR), and the total energy expenditure of the network (COST). We are able to discover network topological configurations over which CTP has unexpectedly low DDR and/or high COST performance, and expose full Pareto fronts which show what the possible performance tradeoffs for CTP are in terms of these two performance factors. Eventually, Pareto fronts allow us to bound the state space of the WSN, a fact which provides essential knowledge to WSN protocol designers.

Defining a distance measure over the individuals in the population of an Evolutionary Algorithm can be exploited for several applications, ranging from diversity preservation to balancing exploration and exploitation. When individuals are encoded as strings of bits or sets of real values, computing the distance between any two can be a straightforward process; when individuals are represented as trees or linear graphs, however, quite often the user must resort to phenotype-level problem-specific distance metrics. This paper presents a generic genotype-level distance metric for Linear Genetic Programming: the information contained by an individual is represented as a set of symbols, using n-grams to capture significant recurring structures inside the genome. The difference in information between two individuals is evaluated resorting to a symmetric difference. Experimental evaluations show that the proposed metric has a strong correlation with phenotype-level problem-specific distance measures in two problems where individuals represent string of bits and Assembly-language programs, respectively.

Abstract-A huge number of optimization problems, in the CAD area as well as in many other fields, require a solution composed by a set of structurally homogeneous elements. Each element tackles a subset of the original task, and they cumulatively solve the whole problem. Sub-tasks, however, have exactly the same structure, and the splitting is completely arbitrary. Even the number of sub-tasks is not known and cannot be determined a-priori. Individual elements are structurally homogeneous, and their contribution to the main solution can be evaluated separately. We propose an evolutionary algorithm able to optimize groups of individuals for solving this class of problems. An individual of the best solution may be sub-optimal when considered alone, but the set of individuals cumulatively represent the optimal group able to completely solve the whole problem. Results of preliminary experiments show that our algorithm performs better than other techniques commonly applied in the CAD field.

Branch Prediction Units (BPUs) are highly efficient modules that can significantly decrease the negative impact of branches in superscalar and RISC processors. Traditional test solutions, mainly based on scan test, are often inadequate to tackle the complexity of these architectures, especially when dealing with delay faults that require at-speed stimuli application. Moreover, scan test does not represent a viable solution when Incoming Inspection or on-line test are considered. In this paper a functional approach targeting BPU test is proposed, allowing to generate a suitable test program whose effectiveness is independent on the specific implementation of the BPU. The effectiveness of the approach is validated on a Branch History Table (BHT) resorting to an open-source computer architecture simulator and to an ad hoc developed HDL testbench. Experimental results show that the proposed method is able to thoroughly test the BHT, reaching complete static fault coverage.

The incessant progress in manufacturing technology is posing new challenges to microprocessor designers. Several activities that were originally supposed to be part of the pre-silicon design phase are migrating after tape-out, when the first silicon prototypes are available. The paper describes a post-silicon methodology for devising functional failing tests. Therefore, suited to be exploited by microprocessor producer to detect, analyze and debug speed paths during verification, speed-stepping, or other critical activities. The proposed methodology is based on an evolutionary algorithm and exploits a versatile toolkit named μGP. The paper describes how to take into account complex hardware characteristics and architectural details of such complex devices. The experimental evaluation clearly demonstrates the potential of this line of research.

Gas chemical sensors are strongly affected by the so-called drift, i.e., changes in sensors' response caused by poisoning and aging that may significantly spoil the measures gathered. The paper presents a mechanism able to correct drift, that is: delivering a correct unbiased fingerprint to the end user. The proposed system exploits a state-of-the-art evolutionary strategy to iteratively tweak the coefficients of a linear transformation. The system operates continuously. The optimal correction strategy is learnt without a-priori models or other hypothesis on the behavior of physical-chemical sensors. Experimental results demonstrate the efficacy of the approach on a real problem.

Caches are crucial components in modern processors (both stand-alone or integrated into SoCs) and their test is a challenging task, especially when addressing complex and high-frequency devices. While the test of the memory array within the cache is usually accomplished resorting to BIST circuitry implementing March test inspired solutions, testing the cache controller logic poses some specific issues, mainly stemming from its limited accessibility. One possible solution consists in letting the processor execute suitable test programs, allowing the detection of possible faults by looking at the results they produce. In this paper we face the issue of generating suitable programs for testing the replacement logic in set-associative caches that implement a deterministic replacement policy. A test program generation approach based on modeling the replacement mechanism as a finite state machine (FSM) is proposed. Experimental results with a cache implementing a LRU policy are provided to assess the effectiveness of the method.

Mobile phones are becoming more and more complex devices, both from the hardware and from the software point of view. Consequently, their various parts are often developed separately. Each sub-system or application may be worked out by a specialized team of engineers and programmers. Frequently, bugs in one component are triggered by the complex interaction between the different applications. Those errors sometimes lead to power dissipation and other misbehaviors that lower residual battery life, a catastrophic event from the user perspective. In this paper we propose a model-based automatic approach to uncover software bugs, which is intended to complement human expertise and complete a qualifying verification plan. The system has been applied on the prototype of a Motorola mobile phone during a partnership with Politecnico di Torino. We demonstrate that our approach is effective by detecting three distinct software misbehaviours that escape all traditional tests. The paper details the methodology, tests and results.