31 research outputs found
Designing Cybersecurity Measurement Systems for Global and Organizational Intelligence
No full textThe growing interdependence of digital infrastructures has expanded organizational
attack surfaces beyond traditional perimeters. This thesis tackles two complementary
problems with distinct methods: (i) generating Cyber Threat Intelligence (CTI) from DNS
cache snooping, where non-recursive queries to public resolvers reveal privacy-preserving
lower bounds on domain interest at global scale; and (ii) maintaining an always-current
view of external exposure by continuously discovering, contextualizing, and prioritizing
Internet-facing assets.
The first contribution, MudHunter, presents a distributed domain name system
(DNS) measurement framework that leverages cache-snooping to infer lower bounds on
domain access activity. By issuing non-recursive queries from 130 globally distributed van-
tage points, MudHunter estimates population-level domain interest without compromising
privacy or requiring authoritative visibility. The resulting empirical results reveal global
access behaviors, regional exposure trends, and malicious ecosystem signals, demonstrating
how passive DNS observation can inform CTI at scale.
The second contribution, the Continuous Threat Exposure Management (CTEM)
framework, operationalizes continuous external risk monitoring. It automates asset discov-
ery, vulnerability enrichment, and risk prioritization into a unified, data-driven pipeline.
The framework integrates large-scale scanning, correlation with structured vulnerability
sources (NVD, CISA KEV, EPSS), and dynamic exposure scoring to provide an always-
current view of organizational risk. A modular architecture, built around event buses, a
database, and RESTful APIs, supports continuous ingestion, enrichment, and visualization
through dashboards and automated interfaces.
viBoth systems share a unifying philosophy: meaningful security insight emerges
from continuous, measurement-based CTI. MudHunter embodies this principle by trans-
forming large-scale DNS cache observations into reproducible empirical evidence about
how global resolvers operate and how malicious infrastructure propagates through them.
CTEM, in turn, applies the same philosophy within organizational environments, continu-
ously measuring, enriching, and prioritizing security exposures through data-driven anal-
ysis. Together, these works advance the state of empirical cyber threat intelligence by
demonstrating that rigorous, measurement-based methodologies can yield deeper under-
standing and more transparent reasoning about the evolving threat landscape
Citizenship struggles: 25th anniversary special issue
No full text© 2022, Informa UK Limited. The attached document (embargoed until 10/01/2024) is an author produced version of a paper published in CITIZENSHIP STUDIES uploaded in accordance with the publisher’s self-archiving policy. The final published version (version of record) is available online at the link. Some minor differences between this version and the final published version may remain. We suggest you refer to the final published version should you wish to cite from it
The code to continuous improvement : a systematic literature review of critical success factors, challenges and tools
No full textLAUREA MAGISTRALEObiettivo: Il miglioramento continuo è una delle implementazioni di maggior impatto per le aziende per far avanzare le proprie operazioni. Oltre ai vantaggi, le aziende soffrono per il mancato raggiungimento dei risultati attesi. Molti autori hanno condotto numerose ricerche sul successo, le sfide, i fallimenti e gli ostacoli al miglioramento continuo. Inoltre, ci sono studi per migliorare i processi di attuazione dei progetti di miglioramento continuo. Tuttavia, è necessario fornire una tabella di marcia teorica che includa fattori critici di successo e sfide piuttosto che una guida pratica all'attuazione.
Design/metodologia/approccio: In questo studio viene adottata una metodologia sistematica di revisione della letteratura sui fattori critici di successo, le sfide e gli strumenti per fornire una tabella di marcia olistica e una visione migliore. Dopo la revisione della letteratura, lo studio ha presentato una tabella di marcia filosofica per le aziende per prepararsi meglio a progetti di successo e risultati superiori.
Risultati: I risultati dello studio elencano i fattori critici di successo, le sfide e gli strumenti più frequentemente citati per il miglioramento continuo con una tabella di marcia teorica a supporto del successo delle iniziative di miglioramento continuo delle aziende.
Limitazioni/implicazioni della ricerca: Limitazioni generali possono essere elencate come la selezione del database Scopus, la lingua inglese, il tipo di articolo cartaceo e, infine, la soggettività del titolo e la raccolta di abstract dell'autore. Tuttavia, non vi è alcun limite di tempo considerato in questo studio.
Implicazioni pratiche: Questo studio può essere utilizzato come preparazione teorica per il miglioramento continuo. Prima di passare alle iniziative di miglioramento continuo nelle aziende, i manager dovrebbero rivedere questa tabella di marcia e assicurarsi che includano tutti i fattori critici di successo ed evitino le sfide per ottenere risultati eccezionali.Purpose: Continuous improvement is one of the most impactful implementations for companies to advance their operations. Besides the advantages, companies suffer from failure in achieving the expected results. Many authors have researched the success, challenges, failures, and barriers to continuous improvement. Additionally, there are studies to improve the implementation processes of continuous improvement projects. However, there is a need to provide a theoretical roadmap that includes critical success factors and challenges besides a practical implementation guide.
Design/methodology/approach: This study adopts the systematic literature review methodology on critical success factors, challenges, and tools to provide a holistic roadmap and better view. After the literature review, the study presented a philosophical roadmap for companies to better prepare for successful projects and superior results.
Findings: The study's outcomes list the most frequently mentioned critical success factors, challenges, and tools for continuous improvement with a theoretical roadmap supporting the success of companies' continuous improvement initiatives.
Research limitations/implications: General limitations can be listed as the selection of the database Scopus, language English, paper type article, and finally, the subjectivity of the title and abstract collection of the author. However, there is no time limit considered in this study.
Practical implications: This study can be used as a theoretical preparation for continuous improvement. Before heading to the continuous improvement initiatives in companies, managers should review this roadmap and make sure that they include all the critical success factors and avoid the challenges to achieve exceptional results
Measuring intercellular interface area in plant tissues using quantitative 3D image analysis
No full textThe cells which make up plant tissues remain fixed together through shared cell walls. Cell-to-cell communication principally takes place through these shared interfaces through a combination of plasmodesmata, transporters, and the apoplastic space. To better understand the capacity for intercellular communication in plant tissues, this chapter outlines a method which can be used to quantify the surface area of shared intercellular interfaces using whole mount imaging and quantitative 3D image analysis. This method allows the potential for intercellular communication as prescribed by cellular architecture to be measured at single cell resolution. [Abstract copyright: © 2022. The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature.
Computational genomics methods to probe cellular and microbial heterogeneity and interactions in cancer
No full textCancer is the result of genomic alterations that drive malignant cell growth in an environment that permits their flourishing and immune evasion. It is a disease in multicellular organism in which individual cells successively devolve into selfish cells with abnormal phenotypes and interactions with their environments. The heterogeneity in tumor microenvironments, which includes varying local physical properties, arrangements of normal and immune cells, and resident or colonizing microbiome, coupled with the diversity of genomic drivers and the co-evolution of this entire system, makes cancer a highly personalized and complex disease. A better understanding of the components in tumors and their interactions is critical to the design of new treatments and could aid in the clinical management of cancer patients. Next-generation sequencing technologies have enabled the widespread profiling of tumors at various scales. In this dissertation, I develop and apply three main computational methods to analyze heterogeneities and interactions in cancer primarily using single-cell RNA-sequencing (scRNA-seq) data.
I developed Census, a biologically intuitive and fully automated cell-type identification method for scRNA-seq data that can deeply annotate normal cells in mammalian tissues and identify malignant cells and their likely cell of origin. Motivated by the inherently stratified developmental programs of cellular differentiation, Census infers hierarchical cell-type relationships and uses gradient-boosted decision trees that capitalize on nodal cell-type relationships to achieve high prediction speed and accuracy. When benchmarked on 44 atlas-scale normal and cancer, human and mouse tissues, Census significantly outperforms state-of-the-art methods across multiple metrics and naturally predicts the cell-of-origin of different cancers. I use Census to identify disseminated pancreatic tumors cells in liver samples without evidence of metastasis and we characterize this rare cell-type.
Cell-cell interactions are the fundamental building blocks of tissue organization and multicellular life. I developed Neighbor-seq, a method to identify the architecture of direct cell-cell interactions and relevant ligand-receptor signaling from the undissociated cell fractions in scRNA-seq data. Neighbor-seq accurately identifies microanatomical features of diverse tissue types such as the small intestinal epithelium, terminal respiratory tract, and splenic white pulp. It also captures the differing topologies of cancer-immune-stromal cell communications in pancreatic and skin tumors, which are consistent with the patterns observed in spatial transcriptomic data. Neighbor-seq provides a framework to study the organ-level cellular interactome in health and disease, bridging the gap between single-cell and spatial transcriptomics.
Lastly, I developed SAHMI, a computational resource to identify truly present microbial nucleic acids and filter contaminants and spurious false-positive taxonomic assignments from standard transcriptomic sequencing of mammalian tissues. In benchmark studies, SAHMI correctly identifies known microbial infections present in diverse tissues. I then use SAHMI to interrogate tumor-microbiome interactions in two human pancreatic cancer cohorts. I identify somatic-cell associated bacteria in a subset of tumors and their near absence in nonmalignant tissues. These bacteria predominantly pair with tumor cells, and their presence associates with cell-type specific gene expression and pathway activities, including cell motility and immune signaling. Modeling results indicate that tumor-infiltrating lymphocytes closely resemble T-cells from infected tissues. Finally, using multiple independent datasets, a signature of cell-associated bacteria predicts clinical prognosis.
Collectively, the algorithms I developed shed light on cancer as a multicellular disease with complex cell-cell and cell-microbe interactions. These tools probe single-cell genomic data at three levels: the identity of individual cells, direct cell-cell interactions, and the recovery of microbial content and its association with host cells. I apply these to find subsets of pancreatic cancer with varying immune subtypes, microorganisms, and survival statistics. My work deepens our understanding of pancreatic cancer specifically, and my studies more generally advance methods to extract signal from complex and noisy biological datasets.Ph.D.Includes bibliographical reference
Regularizing action policies for smooth control with reinforcement learning
No full textA critical problem with the practical utility of
controllers trained with deep Reinforcement Learning (RL)
is the notable lack of smoothness in the actions learned by
the RL policies. This trend often presents itself in the form
of control signal oscillation and can result in poor control,
high power consumption, and undue system wear. We introduce
Conditioning for Action Policy Smoothness (CAPS), an effective
yet intuitive regularization on action policies, which offers consistent
improvement in the smoothness of the learned state-toaction
mappings of neural network controllers, reflected in the
elimination of high-frequency components in the control signal.
Tested on a real system, improvements in controller smoothness
on a quadrotor drone resulted in an almost 80% reduction
in power consumption while consistently training flight-worthy
controllers.First author draf
Author response: KLHL41 stabilizes skeletal muscle sarcomeres by nonproteolytic ubiquitination
No full textSyria's predicament : state (de-) formation and international rivalries
No full textSyria’s war raises important questions about the interaction between the domestic and external dimensions of the conflict. What are the main areas of contention, and how do they relate to regional and international dynamics? Why has the conflict developed into a regional and international battle, and who are the main actors in this rivalry? And, finally, what are the realistic options for ending the Syrian war? The aim of this paper is to answer these questions. In the first, the author examines the domestic origins of the Syrian crisis by focusing on the process of state formation and deformation in Syria. Then, he considers the main areas of contention that shape the Syrian civil war and its regional and international dimensions. Finally, he assesses the conditions under which Syria – as a divided state in a polarised region – can end the war. He argues that in the absence of a military solution to the war in Syria, a political solution may be the only hope for ending the crisis; but such a solution is fraught by varying domestic and external interests in Syria.Peer reviewe
How to train your quadrotor: a framework for consistently smooth and responsive flight control via reinforcement learning
No full textWe focus on the problem of reliably training Reinforcement Learning (RL) models (agents) for stable low-level
control in embedded systems and test our methods on a high-performance, custom-built quadrotor platform.
A common but often under-studied problem in developing RL agents for continuous control is that the control
policies developed are not always smooth. This lack of smoothness can be a major problem when learning
controllers as it can result in control instability and hardware failure.
Issues of noisy control are further accentuated when training RL agents in simulation due to simulators
ultimately being imperfect representations of reality — what is known as the reality gap. To combat issues
of instability in RL agents, we propose a systematic framework, ‘REinforcement-based transferable Agents
through Learning’ (RE+AL), for designing simulated training environments which preserve the quality of
trained agents when transferred to real platforms. RE+AL is an evolution of the Neuroflight infrastructure
detailed in technical reports prepared by members of our research group. Neuroflight is a state-of-the-art
framework for training RL agents for low-level attitude control. RE+AL improves and completes Neuroflight
by solving a number of important limitations that hindered the deployment of Neuroflight to real hardware.
We benchmark RE+AL on the NF1 racing quadrotor developed as part of Neuroflight. We demonstrate that
RE+AL significantly mitigates the previously observed issues of smoothness in RL agents. Additionally, RE+AL
is shown to consistently train agents that are flight-capable and with minimal degradation in controller quality
upon transfer. RE+AL agents also learn to perform better than a tuned PID controller, with better tracking
errors, smoother control and reduced power consumption. To the best of our knowledge, RE+AL agents are
the first RL-based controllers trained in simulation to outperform a well-tuned PID controller on a real-world
controls problem that is solvable with classical control.First author draf
