Alex Serban

Alex Serban, Magiel Bruntink, Joost Visser

Mining and storage of data from software repositories is typically done on a per-project basis, where each project uses a unique combination of data schema, extraction tools, and (intermediate) storage infrastructure. We introduce GraphRepo, a tool that enables a unified approach to extract data from Git repositories, store it, and share it across repository mining projects. GraphRepo usesNeo4j, an ACID-compliant graph database management system, and allows modular plug-in of components for repository extraction (drillers), analysis (miners), and export (mappers). The graph enables a natural way to query the data by removing the need for data normalisation. GraphRepo is built in Python and offers multiple ways to interface with the rich Python ecosystem and with big data solutions. The schema of the graph database is generic and extensible. Using GraphRepo for software repository mining offers several advantages versus creating project-specific infrastructure: (i) high performance for short-iteration exploration and scalability to large data sets (ii) easy distribution of extracted data(e.g., for replication) or sharing of extracted data among projects, and (iii) extensibility and interoperability. A set of benchmarks on four open source projects demonstrate that GraphRepo allows very fast querying of repository data, once extracted and indexed. More information can be found in the project's documentation (available at https://tinyurl.com/grepodoc) and in the project's repository (available at https://tinyurl.com/grrepo). A video demonstration isalso available online (https://tinyurl.com/grrepov)

Andrei Gasparovici, Alex Serban

We propose a deep learning method to model and generate synthetic aortic shapes based on representing shapes as the zero-level set of a neural signed distance field, conditioned by a family of trainable embedding vectors with encode the geometric features of each shape. The network is trained on a dataset of aortic root meshes reconstructed from CT images by making the neural field vanish on sampled surface points and enforcing its spatial gradient to have unit norm. Empirical results show that our model can represent aortic shapes with high fidelity. Moreover, by sampling from the learned embedding vectors, we can generate novel shapes that resemble real patient anatomies, which can be used for in-silico trials.

Costin F. Ciusdel, Alex Serban, Tiziano Passerini

While traditional self-supervised learning methods improve performance and robustness across various medical tasks, they rely on single-vector embeddings that may not capture fine-grained concepts such as anatomical structures or organs. The ability to identify such concepts and their characteristics without supervision has the potential to improve pre-training methods, and enable novel applications such as fine-grained image retrieval and concept-based outlier detection. In this paper, we introduce ConceptVAE, a novel pre-training framework that detects and disentangles fine-grained concepts from their style characteristics in a self-supervised manner. We present a suite of loss terms and model architecture primitives designed to discretise input data into a preset number of concepts along with their local style. We validate ConceptVAE both qualitatively and quantitatively, demonstrating its ability to detect fine-grained anatomical structures such as blood pools and septum walls from 2D cardiac echocardiographies. Quantitatively, ConceptVAE outperforms traditional self-supervised methods in tasks such as region-based instance retrieval, semantic segmentation, out-of-distribution detection, and object detection. Additionally, we explore the generation of in-distribution synthetic data that maintains the same concepts as the training data but with distinct styles, highlighting its potential for more calibrated data generation. Overall, our study introduces and validates a promising new pre-training technique based on concept-style disentanglement, opening multiple avenues for developing models for medical image analysis that are more interpretable and explainable than black-box approaches.

Cosmin-Andrei Hatfaludi, Alex Serban

Federated learning has the potential to unlock siloed data and distributed resources by enabling collaborative model training without sharing private data. As more complex foundational models gain widespread use, the need to expand training resources and integrate privately owned data grows as well. In this article, we explore the intersection of federated learning and foundational models, aiming to identify, categorize, and characterize technical methods that integrate the two paradigms. As a unified survey is currently unavailable, we present a literature survey structured around a novel taxonomy that follows the development life-cycle stages, along with a technical comparison of available methods. Additionally, we provide practical insights and guidelines for implementing and evolving these methods, with a specific focus on the healthcare domain as a case study, where the potential impact of federated learning and foundational models is considered significant. Our survey covers multiple intersecting topics, including but not limited to federated learning, self-supervised learning, fine-tuning, distillation, and transfer learning. Initially, we retrieved and reviewed a set of over 4,200 articles. This collection was narrowed to more than 250 thoroughly reviewed articles through inclusion criteria, featuring 42 unique methods. The methods were used to construct the taxonomy and enabled their comparison based on complexity, efficiency, and scalability. We present these results as a self-contained overview that not only summarizes the state of the field but also provides insights into the practical aspects of adopting, evolving, and integrating foundational models with federated learning.

Cosmin-Andrei Hatfaludi, Daniel Bunescu, Costin Florian Ciusdel et al.

Coronary artery disease (CAD) is the dominant cause of death and hospitalization across the globe. Atherosclerosis, an inflammatory condition that gradually narrows arteries and has potentially fatal effects, is the most frequent cause of CAD. Nonetheless, the circulation regularly adapts in the presence of atherosclerosis, through the formation of collateral arteries, resulting in significant long-term health benefits. Therefore, timely detection of coronary collateral circulation (CCC) is crucial for CAD personalized medicine. We propose a novel deep learning based method to detect CCC in angiographic images. Our method relies on a convolutional backbone to extract spatial features from each frame of an angiography sequence. The features are then concatenated, and subsequently processed by another convolutional layer that processes embeddings temporally. Due to scarcity of data, we also experiment with pretraining the backbone on coronary artery segmentation, which improves the results consistently. Moreover, we experiment with few-shot learning to further improve performance, given our low data regime. We present our results together with subgroup analyses based on Rentrop grading, collateral flow, and collateral grading, which provide valuable insights into model performance. Overall, the proposed method shows promising results in detecting CCC, and can be further extended to perform landmark based CCC detection and CCC quantification.

Alex Serban, Erik Poll, Joost Visser

While many defences against adversarial examples have been proposed, finding robust machine learning models is still an open problem. The most compelling defence to date is adversarial training and consists of complementing the training data set with adversarial examples. Yet adversarial training severely impacts training time and depends on finding representative adversarial samples. In this paper we propose to train models on output spaces with large class separation in order to gain robustness without adversarial training. We introduce a method to partition the output space into class prototypes with large separation and train models to preserve it. Experimental results shows that models trained with these prototypes -- which we call deep repulsive prototypes -- gain robustness competitive with adversarial training, while also preserving more accuracy on natural samples. Moreover, the models are more resilient to large perturbation sizes. For example, we obtained over 50% robustness for CIFAR-10, with 92% accuracy on natural samples and over 20% robustness for CIFAR-100, with 71% accuracy on natural samples without adversarial training. For both data sets, the models preserved robustness against large perturbations better than adversarially trained models.

Alex Serban, Joost Visser

Unique developmental and operational characteristics of ML components as well as their inherent uncertainty demand robust engineering principles are used to ensure their quality. We aim to determine how software systems can be (re-) architected to enable robust integration of ML components. Towards this goal, we conducted a mixed-methods empirical study consisting of (i) a systematic literature review to identify the challenges and their solutions in software architecture for ML, (ii) semi-structured interviews with practitioners to qualitatively complement the initial findings and (iii) a survey to quantitatively validate the challenges and their solutions. We compiled and validated twenty challenges and solutions for (re-) architecting systems with ML components. Our results indicate, for example, that traditional software architecture challenges (e.g., component coupling) also play an important role when using ML components; along with new ML specific challenges (e.g., the need for continuous retraining). Moreover, the results indicate that ML heightened decision drivers, such as privacy, play a marginal role compared to traditional decision drivers, such as scalability. Using the survey we were able to establish a link between architectural solutions and software quality attributes, which enabled us to provide twenty architectural tactics used to satisfy individual quality requirements of systems with ML components. Altogether, the results of the study can be interpreted as an empirical framework that supports the process of (re-) architecting software systems with ML components.

Alex Serban, Koen van der Blom, Holger Hoos et al.

Following the recent surge in adoption of machine learning (ML), the negative impact that improper use of ML can have on users and society is now also widely recognised. To address this issue, policy makers and other stakeholders, such as the European Commission or NIST, have proposed high-level guidelines aiming to promote trustworthy ML (i.e., lawful, ethical and robust). However, these guidelines do not specify actions to be taken by those involved in building ML systems. In this paper, we argue that guidelines related to the development of trustworthy ML can be translated to operational practices, and should become part of the ML development life cycle. Towards this goal, we ran a multi-vocal literature review, and mined operational practices from white and grey literature. Moreover, we launched a global survey to measure practice adoption and the effects of these practices. In total, we identified 14 new practices, and used them to complement an existing catalogue of ML engineering practices. Initial analysis of the survey results reveals that so far, practice adoption for trustworthy ML is relatively low. In particular, practices related to assuring security of ML components have very low adoption. Other practices enjoy slightly larger adoption, such as providing explanations to users. Our extended practice catalogue can be used by ML development teams to bridge the gap between high-level guidelines and actual development of trustworthy ML systems; it is open for review and contribution

Alex Serban, Erik Poll, Joost Visser

Sensitivity to adversarial noise hinders deployment of machine learning algorithms in security-critical applications. Although many adversarial defenses have been proposed, robustness to adversarial noise remains an open problem. The most compelling defense, adversarial training, requires a substantial increase in processing time and it has been shown to overfit on the training data. In this paper, we aim to overcome these limitations by training robust models in low data regimes and transfer adversarial knowledge between different models. We train a meta-optimizer which learns to robustly optimize a model using adversarial examples and is able to transfer the knowledge learned to new models, without the need to generate new adversarial examples. Experimental results show the meta-optimizer is consistent across different architectures and data sets, suggesting it is possible to automatically patch adversarial vulnerabilities.

Alex Serban, Erik Poll, Joost Visser

The adoption of machine learning (ML) components in software systems raises new engineering challenges. In particular, the inherent uncertainty regarding functional suitability and the operation environment makes architecture evaluation and trade-off analysis difficult. We propose a software architecture evaluation method called Modeling Uncertainty During Design (MUDD) that explicitly models the uncertainty associated to ML components and evaluates how it propagates through a system. The method supports reasoning over how architectural patterns can mitigate uncertainty and enables comparison of different architectures focused on the interplay between ML and classical software components. While our approach is domain-agnostic and suitable for any system where uncertainty plays a central role, we demonstrate our approach using as example a perception system for autonomous driving.

Alex Serban, Erik Poll, Joost Visser

Deep neural networks are at the forefront of machine learning research. However, despite achieving impressive performance on complex tasks, they can be very sensitive: Small perturbations of inputs can be sufficient to induce incorrect behavior. Such perturbations, called adversarial examples, are intentionally designed to test the network's sensitivity to distribution drifts. Given their surprisingly small size, a wide body of literature conjectures on their existence and how this phenomenon can be mitigated. In this article we discuss the impact of adversarial examples on security, safety, and robustness of neural networks. We start by introducing the hypotheses behind their existence, the methods used to construct or protect against them, and the capacity to transfer adversarial examples between different machine learning models. Altogether, the goal is to provide a comprehensive and self-contained survey of this growing field of research.

Alex Serban, Koen van der Blom, Holger Hoos et al.

The increasing reliance on applications with machine learning (ML) components calls for mature engineering techniques that ensure these are built in a robust and future-proof manner. We aim to empirically determine the state of the art in how teams develop, deploy and maintain software with ML components. We mined both academic and grey literature and identified 29 engineering best practices for ML applications. We conducted a survey among 313 practitioners to determine the degree of adoption for these practices and to validate their perceived effects. Using the survey responses, we quantified practice adoption, differentiated along demographic characteristics, such as geography or team size. We also tested correlations and investigated linear and non-linear relationships between practices and their perceived effect using various statistical models. Our findings indicate, for example, that larger teams tend to adopt more practices, and that traditional software engineering practices tend to have lower adoption than ML specific practices. Also, the statistical models can accurately predict perceived effects such as agility, software quality and traceability, from the degree of adoption for specific sets of practices. Combining practice adoption rates with practice importance, as revealed by statistical models, we identify practices that are important but have low adoption, as well as practices that are widely adopted but are less important for the effects we studied. Overall, our survey and the analysis of responses received provide a quantitative basis for assessment and step-wise improvement of practice adoption by ML teams.