Efficient energy provisioning is a fundamental requirement for modern transportation systems, making refueling path optimization a critical challenge. Existing solutions often focus either on inter-vehicle communication or intra-vehicle monitoring, leveraging Intelligent Transportation Systems, Digital Twins, and Software-Defined Internet of Vehicles with Cloud/Fog/Edge infrastructures. However, integrated frameworks that adapt dynamically to driver mobility patterns are still underdeveloped. Building on our previous PIENO framework, we present RI-PIENO (Revised and Improved Petrol-filling Itinerary Estimation aNd Optimization), a system that combines intra-vehicle sensor data with external geospatial and fuel price information, processed via IoT-enabled Cloud/Fog services. RI-PIENO models refueling as a dynamic, time-evolving directed acyclic graph that reflects both habitual daily trips and real-time vehicular inputs, transforming the system from a static recommendation tool into a continuously adaptive decision engine. We validate RI-PIENO in a daily-commute use case through realistic multi-driver, multi-week simulations, showing that it achieves significant cost savings and more efficient routing compared to previous approaches. The framework is designed to leverage emerging roadside infrastructure and V2X communication, supporting scalable deployment within next-generation IoT and vehicular networking ecosystems.

The complete workflow of the RI-PIENO framework is presented, a system for refueling itinerary optimization that extends the original PIENO design. While prior work introduced the conceptual modules of RI-PIENO, their operational pipeline was not described in detail. This study makes the workflow explicit, covering the end-to-end process from CAN Bus data acquisition and stop detection to the construction of daily trip graphs, refueling optimization, and mileage prediction. By clarifying the sequence of operations, the contribution provides a reproducible and extensible foundation for future research and development.

Generating faithful and readable styled text images (especially for Styled Handwritten Text generation-HTG) is an open problem with several possible applications across graphic design, document understanding, and image editing. A lot of research effort in this task is dedicated to developing strategies that reproduce the stylistic characteristics of a given writer, with promising results in terms of style fidelity and generalization achieved by the recently proposed Autoregressive Transformer paradigm for HTG. However, this method requires additional inputs, lacks a proper stop mechanism, and might end up in repetition loops, generating visual artifacts. In this work, we rethink the autoregressive formulation by framing HTG as a multimodal prompt-conditioned generation task, and tackle the content controllability issues by introducing special textual input tokens for better alignment with the visual ones. Moreover, we devise a Classifier-Free-Guidance-based strategy for our autoregressive model. Through extensive experimental validation, we demonstrate that our approach, dubbed Eruku, compared to previous solutions requires fewer inputs, generalizes better to unseen styles, and follows more faithfully the textual prompt, improving content adherence.

Styled Handwritten Text Generation (HTG) has recently received attention from the computer vision and document analysis communities, which have developed several solutions, either GAN- or diffusion-based, that achieved promising results. Nonetheless, these strategies fail to generalize to novel styles and have technical constraints, particularly in terms of maximum output length and training efficiency. To overcome these limitations, in this work, we propose a novel framework for text image generation, dubbed Emuru. Our approach leverages a powerful text image representation model (a variational autoencoder) combined with an autoregressive Transformer. Our approach enables the generation of styled text images conditioned on textual content and style examples, such as specific fonts or handwriting styles. We train our model solely on a diverse, synthetic dataset of English text rendered in over 100,000 typewritten and calligraphy fonts, which gives it the capability to reproduce unseen styles (both fonts and users’ handwriting) in zero-shot. To the best of our knowledge, Emuru is the first autoregressive model for HTG, and the first designed specifically for generalization to novel styles. Moreover, our model generates images without background artifacts, which are easier to use for downstream applications. Extensive evaluation on both typewritten and handwritten, anylength text image generation scenarios demonstrates the effectiveness of our approach.

OBJECTIVES: Patient classification into specific molecular subtypes is paramount in biomedical research and clinical practice to face complex, heterogeneous diseases. Existing methods, especially for gene expression-based cancer subtyping, often simplify patient molecular portraits, neglecting the potential co-occurrence of traits from multiple subtypes. Yet, recognizing intra-sample heterogeneity is essential for more precise patient characterization and improved personalized treatments. METHODS: We developed a novel computational workflow, named MULTI-STAR, which addresses current limitations and provides tailored solutions for reliable multi-label patient subtyping. MULTI-STAR uses state-of-the-art subtyping methods to obtain promising machine learning-based multi-label classifiers, leveraging gene expression profiles. It modifies standard single-label similarity-based techniques to obtain multi-label patient characterizations. Then, it employs these characterizations to train single-sample predictors using different multi-label strategies and find the best-performing classifiers. RESULTS: MULTI-STAR classifiers offer advanced multi-label recognition of all the subtypes contributing to the molecular and clinical traits of a patient, also distinguishing the primary from the additional relevant secondary subtype(s). The efficacy was demonstrated by developing multi-label solutions for breast and colorectal cancer subtyping that outperform existing methods in terms of prognostic value, primarily for overall survival predictions, and ability to work on a single sample at a time, as required in clinical practice. CONCLUSIONS: This work emphasizes the importance of moving to multi-label subtyping to capture all the molecular traits of individual patients, considering also previously overlooked secondary assignments and paving the way for improved clinical decision-making processes in diverse heterogeneous disease contexts. Indeed, MULTI-STAR novel, reproducible and generalizable approach provides comprehensive representations of patient inner heterogeneity and clinically relevant insights, contributing to precision medicine and personalized treatments.

Styled Handwritten Text Generation (HTG) has received significant attention in recent years, propelled by the success of learning-based solutions employing GANs, Transformers, and, preliminarily, Diffusion Models. Despite this surge in interest, there remains a critical yet understudied aspect - the impact of the input, both visual and textual, on the HTG model training and its subsequent influence on performance. This work extends the VATr (Pippi et al. 2023) Styled-HTG approach by addressing the pre-processing and training issues that it faces, which are common to many HTG models. In particular, we propose generally applicable strategies for input preparation and training regularization that allow the model to achieve better performance and generalization capabilities. Moreover, in this work, we go beyond performance optimization and address a significant hurdle in HTG research - the lack of a standardized evaluation protocol. In particular, we propose a standardization of the evaluation protocol for HTG and conduct a comprehensive benchmarking of existing approaches. By doing so, we aim to establish a foundation for fair and meaningful comparisons between HTG strategies, fostering progress in the field.

MOTIVATION: Feature selection approaches are widely used in gene expression data analysis to identify the most relevant features and boost performance in regression and classification tasks. However, such algorithms solely consider each feature's quantitative contribution to the task, possibly limiting the biological interpretability of the results. Feature-related prior knowledge, such as functional annotations and pathways information, can be incorporated into feature selection algorithms to potentially improve model performance and interpretability. RESULTS: We propose an embedded integrative approach to feature selection that combines weighted LASSO feature selection and prior biological knowledge in a single step, by means of a novel score of biological relevance that summarizes information extracted from popular biological knowledge bases. Findings from the performed experiments indicate that our proposed approach is able to identify the most predictive genes while simultaneously enhancing the biological interpretability of the results compared to the standard LASSO regularized model. AVAILABILITY AND IMPLEMENTATION: Code is available at https://github.com/DEIB-GECO/GIS-weigthed_LASSO.

Generating synthetic images of handwritten text in a writer-specific style is a challenging task, especially in the case of unseen styles and new words, and even more when these latter contain characters that are rarely encountered during training. While emulating a writer's style has been recently addressed by generative models, the generalization towards rare characters has been disregarded. In this work, we devise a Transformer-based model for Few-Shot styled handwritten text generation and focus on obtaining a robust and informative representation of both the text and the style. In particular, we propose a novel representation of the textual content as a sequence of dense vectors obtained from images of symbols written as standard GNU Unifont glyphs, which can be considered their visual archetypes. This strategy is more suitable for generating characters that, despite having been seen rarely during training, possibly share visual details with the frequently observed ones. As for the style, we obtain a robust representation of unseen writers' calligraphy by exploiting specific pre-training on a large synthetic dataset. Quantitative and qualitative results demonstrate the effectiveness of our proposal in generating words in unseen styles and with rare characters more faithfully than existing approaches relying on independent one-hot encodings of the characters.

The development of embodied agents that can communicate with humans in natural language has gained increasing interest over the last years, as it facilitates the diffusion of robotic platforms in human-populated environments. As a step towards this objective, in this work, we tackle a setting for visual navigation in which an autonomous agent needs to explore and map an unseen indoor environment while portraying interesting scenes with natural language descriptions. To this end, we propose and evaluate an approach that combines recent advances in visual robotic exploration and image captioning on images generated through agent-environment interaction. Our approach can generate smart scene descriptions that maximize semantic knowledge of the environment and avoid repetitions. Further, such descriptions offer user-understandable insights into the robot's representation of the environment by high-lighting the prominent objects and the correlation between them as encountered during the exploration. To quantitatively assess the performance of the proposed approach, we also devise a specific score that takes into account both exploration and description skills. The experiments carried out on both photorealistic simulated environments and real-world ones demonstrate that our approach can effectively describe the robot's point of view during exploration, improving the human-friendly interpretability of its observations.

BACKGROUND: Transcription factors (TF) play a crucial role in the regulation of gene transcription; alterations of their activity and binding to DNA areas are strongly involved in cancer and other disease onset and development. For proper biomedical investigation, it is hence essential to correctly trace TF dense DNA areas, having multiple bindings of distinct factors, and select DNA high occupancy target (HOT) zones, showing the highest accumulation of such bindings. Indeed, systematic and replicable analysis of HOT zones in a large variety of cells and tissues would allow further understanding of their characteristics and could clarify their functional role. RESULTS: Here, we propose, thoroughly explain and discuss a full computational procedure to study in-depth DNA dense areas of transcription factor accumulation and identify HOT zones. This methodology, developed as a computationally efficient parametric algorithm implemented in an R/Bioconductor package, uses a systematic approach with two alternative methods to examine transcription factor bindings and provide comparative and fully-reproducible assessments. It offers different resolutions by introducing three distinct types of accumulation, which can analyze DNA from single-base to region-oriented levels, and a moving window, which can estimate the influence of the neighborhood for each DNA base under exam. CONCLUSIONS: We quantitatively assessed the full procedure by using our implemented software package, named TFHAZ, in two example applications of biological interest, proving its full reliability and relevance.

BACKGROUND AND OBJECTIVE: Many classification tasks in translational bioinformatics and genomics are characterized by the high dimensionality of potential features and unbalanced sample distribution among classes. This can affect classifier robustness and increase the risk of overfitting, curse of dimensionality and generalization leaks; furthermore and most importantly, this can prevent obtaining adequate patient stratification required for precision medicine in facing complex diseases, like cancer. Setting up a feature selection strategy able to extract only proper predictive features by removing irrelevant, redundant, and noisy ones is crucial to achieving valuable results on the desired task. METHODS: We propose a new feature selection approach, called ReRa, based on supervised Relevance-Redundancy assessments. ReRa consists of a customized step of relevance-based filtering, to identify a reduced subset of meaningful features, followed by a supervised similarity-based procedure to minimize redundancy. This latter step innovatively uses a combination of global and class-specific similarity assessments to remove redundant features while preserving those differentiated across classes, even when these classes are strongly unbalanced. RESULTS: We compared ReRa with several existing feature selection methods to obtain feature spaces on which performing breast cancer patient subtyping using several classifiers: we considered two use cases based on gene or transcript isoform expression. In the vast majority of the assessed scenarios, when using ReRa-selected feature spaces, the performances were significantly increased compared to simple feature filtering, LASSO regularization, or even MRmr - another Relevance-Redundancy method. The two use cases represent an insightful example of translational application, taking advantage of ReRa capabilities to investigate and enhance a clinically-relevant patient stratification task, which could be easily applied also to other cancer types and diseases. CONCLUSIONS: ReRa approach has the potential to improve the performance of machine learning models used in an unbalanced classification scenario. Compared to another Relevance-Redundancy approach like MRmr, ReRa does not require tuning the number of preserved features, ensures efficiency and scalability over huge initial dimensionalities and allows re-evaluation of all previously selected features at each iteration of the redundancy assessment, to ultimately preserve only the most relevant and class-differentiated features.

Connecting Vision and Language plays an essential role in Generative Intelligence. For this reason, large research efforts have been devoted to image captioning, i.e. describing images with syntactically and semantically meaningful sentences. Starting from 2015 the task has generally been addressed with pipelines composed of a visual encoder and a language model for text generation. During these years, both components have evolved considerably through the exploitation of object regions, attributes, the introduction of multi-modal connections, fully-attentive approaches, and BERT-like early-fusion strategies. However, regardless of the impressive results, research in image captioning has not reached a conclusive answer yet. This work aims at providing a comprehensive overview of image captioning approaches, from visual encoding and text generation to training strategies, datasets, and evaluation metrics. In this respect, we quantitatively compare many relevant state-of-the-art approaches to identify the most impactful technical innovations in architectures and training strategies. Moreover, many variants of the problem and its open challenges are discussed. The final goal of this work is to serve as a tool for understanding the existing literature and highlighting the future directions for a research area where Computer Vision and Natural Language Processing can find an optimal synergy.

In this work, we explore massive pre-training on synthetic word images for enhancing the performance on four benchmark downstream handwriting analysis tasks. To this end, we build a large synthetic dataset of word images rendered in several handwriting fonts, which offers a complete supervision signal. We use it to train a simple convolutional neural network (ConvNet) with a fully supervised objective. The vector representations of the images obtained from the pre-trained ConvNet can then be considered as encodings of the handwriting style. We exploit such representations for Writer Retrieval, Writer Identification, Writer Verification, and Writer Classification and demonstrate that our pre-training strategy allows extracting rich representations of the writers’ style that enable the aforementioned tasks with competitive results with respect to task-specific State-of-the-Art approaches.

Describing images in natural language is a fundamental step towards the automatic modeling of connections between the visual and textual modalities. In this paper we present CaMEL, a novel Transformer-based architecture for image captioning. Our proposed approach leverages the interaction of two interconnected language models that learn from each other during the training phase. The interplay between the two language models follows a mean teacher learning paradigm with knowledge distillation. Experimentally, we assess the effectiveness of the proposed solution on the COCO dataset and in conjunction with different visual feature extractors. When comparing with existing proposals, we demonstrate that our model provides state-of-the-art caption quality with a significantly reduced number of parameters. According to the CIDEr metric, we obtain a new state of the art on COCO when training without using external data. The source code and trained models will be made publicly available at: https://github.com/aimagelab/camel.

Handwritten Text Recognition (HTR) is an open problem at the intersection of Computer Vision and Natural Language Processing. The main challenges, when dealing with historical manuscripts, are due to the preservation of the paper support, the variability of the handwriting – even of the same author over a wide time-span – and the scarcity of data from ancient, poorly represented languages. With the aim of fostering the research on this topic, in this paper we present the Ludovico Antonio Muratori (LAM) dataset, a large line-level HTR dataset of Italian ancient manuscripts edited by a single author over 60 years. The dataset comes in two configurations: a basic splitting and a date-based splitting which takes into account the age of the author. The first setting is intended to study HTR on ancient documents in Italian, while the second focuses on the ability of HTR systems to recognize text written by the same writer in time periods for which training data are not available. For both configurations, we analyze quantitative and qualitative characteristics, also with respect to other line-level HTR benchmarks, and present the recognition performance of state-of-the-art HTR architectures. The dataset is available for download at https://aimagelab.ing.unimore.it/go/lam.

Embodied AI is a recent research area that aims at creating intelligent agents that can move and operate inside an environment. Existing approaches in this field demand the agents to act in completely new and unexplored scenes. However, this setting is far from realistic use cases that instead require executing multiple tasks in the same environment. Even if the environment changes over time, the agent could still count on its global knowledge about the scene while trying to adapt its internal representation to the current state of the environment. To make a step towards this setting, we propose Spot the Difference: a novel task for Embodied AI where the agent has access to an outdated map of the environment and needs to recover the correct layout in a fixed time budget. To this end, we collect a new dataset of occupancy maps starting from existing datasets of 3D spaces and generating a number of possible layouts for a single environment. This dataset can be employed in the popular Habitat simulator and is fully compliant with existing methods that employ reconstructed occupancy maps during navigation. Furthermore, we propose an exploration policy that can take advantage of previous knowledge of the environment and identify changes in the scene faster and more effectively than existing agents. Experimental results show that the proposed architecture outperforms existing state-of-the-art models for exploration on this new setting.

BACKGROUND: Heterogeneous omics data, increasingly collected through high-throughput technologies, can contain hidden answers to very important and still unsolved biomedical questions. Their integration and processing are crucial mostly for tertiary analysis of Next Generation Sequencing data, although suitable big data strategies still address mainly primary and secondary analysis. Hence, there is a pressing need for algorithms specifically designed to explore big omics datasets, capable of ensuring scalability and interoperability, possibly relying on high-performance computing infrastructures. RESULTS: We propose RGMQL, a R/Bioconductor package conceived to provide a set of specialized functions to extract, combine, process and compare omics datasets and their metadata from different and differently localized sources. RGMQL is built over the GenoMetric Query Language (GMQL) data management and computational engine, and can leverage its open curated repository as well as its cloud-based resources, with the possibility of outsourcing computational tasks to GMQL remote services. Furthermore, it overcomes the limits of the GMQL declarative syntax, by guaranteeing a procedural approach in dealing with omics data within the R/Bioconductor environment. But mostly, it provides full interoperability with other packages of the R/Bioconductor framework and extensibility over the most used genomic data structures and processing functions. CONCLUSIONS: RGMQL is able to combine the query expressiveness and computational efficiency of GMQL with a complete processing flow in the R environment, being a fully integrated extension of the R/Bioconductor framework. Here we provide three fully reproducible example use cases of biological relevance that are particularly explanatory of its flexibility of use and interoperability with other R/Bioconductor packages. They show how RGMQL can easily scale up from local to parallel and cloud computing while it combines and analyzes heterogeneous omics data from local or remote datasets, both public and private, in a completely transparent way to the user.

Breast Cancer comprises multiple subtypes implicated in prognosis. Existing stratification methods rely on the expression quantification of small gene sets. Next Generation Sequencing promises large amounts of omic data in the next years. In this scenario, we explore the potential of machine learning and, particularly, deep learning for breast cancer subtyping. Due to the paucity of publicly available data, we leverage on pan-cancer and non-cancer data to design semi-supervised settings. We make use of multi-omic data, including microRNA expressions and copy number alterations, and we provide an in-depth investigation of several supervised and semi-supervised architectures. Obtained accuracy results show simpler models to perform at least as well as the deep semi-supervised approaches on our task over gene expression data. When multi-omic data types are combined together, performance of deep models shows little (if any) improvement in accuracy, indicating the need for further analysis on larger datasets of multi-omic data as and when they become available. From a biological perspective, our linear model mostly confirms known gene-subtype annotations. Conversely, deep approaches model non-linear relationships, which is reflected in a more varied and still unexplored set of representative omic features that may prove useful for breast cancer subtyping.

The power produced by a wind turbine depends on environmental conditions, working parameters, and interactions with nearby turbines. However, these aspects are often neglected in the design of data-driven models for wind farms’ performance analysis. In this article, we propose to predict the active power and to provide reliable prediction intervals via ensembles of multivariate polynomial regression models that exploit a higher number of inputs (compared to most approaches in the literature), including operational and thermal variables. We present two main strategies: the former considers the environmental measurements collected at the other wind turbines in the farm as additional modeling information for the turbine under analysis; the latter combines multiple models relative to different operative conditions. We validate our approach on real data from the SCADA system of a wind farm in Italy and obtain a MAE of the order of 1.0% of the rated power of the turbine. Moreover, due to the structure of our approach, we can gain quantitative insights on the covariates most frequently selected depending on the working region of the wind turbines.

Embodied AI has been recently gaining attention as it aims to foster the development of autonomous and intelligent agents. In this paper, we devise a novel embodied setting in which an agent needs to explore a previously unknown environment while recounting what it sees during the path. In this context, the agent needs to navigate the environment driven by an exploration goal, select proper moments for description, and output natural language descriptions of relevant objects and scenes. Our model integrates a novel self-supervised exploration module with penalty, and a fully-attentive captioning model for explanation. Also, we investigate different policies for selecting proper moments for explanation, driven by information coming from both the environment and the navigation. Experiments are conducted on photorealistic environments from the Matterport3D dataset and investigate the navigation and explanation capabilities of the agent as well as the role of their interactions.

Handwritten Text Recognition (HTR) in free-layout pages is a valuable yet challenging task which aims to automatically understand handwritten texts. State-of-the-art approaches in this field usually encode input images with Convolutional Neural Networks, whose kernels are typically defined on a fixed grid and focus on all input pixels independently. However, this is in contrast with the sparse nature of handwritten pages, in which only pixels representing the ink of the writing are useful for the recognition task. Furthermore, the standard convolution operator is not explicitly designed to take into account the great variability in shape, scale, and orientation of handwritten characters. To overcome these limitations, we investigate the use of deformable convolutions for handwriting recognition. The kernel of this type of convolution deforms according to the content of the neighborhood, and can therefore be more adaptable to geometric variations and other deformations of the text. Experiments conducted on the IAM and RIMES datasets demonstrate that the use of deformable convolutions is a promising direction for the design of novel architectures for handwritten text recognition.

Predicting relapse events is still one of the major challenges for breast cancer research. Despite gene expression-based classifiers may tackle this task, working on thousands of genes and only few samples jeopardizes the performances of a classifier trained without a proper gene selection. We propose a novel hybrid evolutionary gene selection framework, which uses a Multi-Objective Genetic Algorithm (MOGA) to search a wider range of gene selections and handles MOGA results in a whole new way, so as to overcome the limit of the non-easy interpretability of the MOGA broad set of solutions. To a classifier our framework provides a gene signature not only bringing the best cross-validation result, but also having noteworthy and robust performances when tested on unseen samples of an hold-out set. The robustness in hold-out showed the strength of our innovative key element: the final module of the framework, which fully exploits the high variability of MOGA outputs, rather than choosing just one of the solutions, as commonly done in the literature. It combines all MOGA results in more robust and compact gene occurrence-based signatures, under the reasonable assumption that highly recurrent genes have a more crucial biological role, more suitable clinical application and good discriminative power between relapsed and relapse-free patients, as confirmed by the obtained classification results.

Natural language video description (NLVD) has recently received strong interest in the computer vision, natural language processing (NLP), multimedia, and autonomous robotics communities. The state-of-the-art (SotA) approaches obtained remarkable results when tested on the benchmark datasets. However, those approaches poorly generalize to new datasets. In addition, none of the existing works focus on the processing of the input to the NLVD systems, which is both visual and textual. In this paper, an extensive study is presented to deal with the role of the visual input, evaluated with respect to the overall NLP performance. This is achieved by performing data augmentation of the visual component, applying common transformations to model camera distortions, noise, lighting, and camera positioning that are typical in real-world operative scenarios. A t-SNE-based analysis is proposed to evaluate the effects of the considered transformations on the overall visual data distribution. For this study, the English subset of the Microsoft Research Video Description (MSVD) dataset is considered, which is used commonly for NLVD. It was observed that this dataset contains a relevant amount of syntactic and semantic errors. These errors have been amended manually, and the new version of the dataset (called MSVD-v2) is used in the experimentation. The MSVD-v2 dataset is released to help to gain insight into the NLVD problem.