liu.seSök publikationer i DiVA
1 - 2 av 2
rss atomLänk till träfflistan
Permanent länk
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
  • Presentation: 2025-12-05 10:15 K2, NorrköpingBeställ onlineKöp publikationen >>
    Berggren, Helene
    Linköpings universitet, Institutionen för beteendevetenskap och lärande, Avdelningen för lärande, estetik och naturvetenskap. Linköpings universitet, Utbildningsvetenskap.
    Att närma sig naturvetenskap: De yngsta barnens utforskande i förskolan2025Licentiatavhandling, monografi (Övrigt vetenskapligt)
    Abstract [sv]

    Denna studie undersöker hur de yngsta barnen (1–3 år) i svenska förskolor möter och engagerar sig i det som kan omvandlas till naturvetenskap genom vardaglig lek och interaktion. Syftet är att generera kunskap om hur begynnande naturvetenskapligt lärande tar form i förskolan och hur barns handlingar, både på egen hand, tillsammans med andra barn och i samspel med pedagoger, kan förstås som meningsskapande processer kopplade till naturvetenskap.

    Det teoretiska ramverket utgår från symbolisk interaktionism, vilket synliggör hur mening skapas genom social interaktion och användningen av sociala objekt. Metodologiskt vilar studien på en konstruktivistisk grundad teori-ansats, baserad på etnografiskt fältarbete med videoobservationer, fältanteckningar och informella samtal i flera förskolemiljöer.

    Analysen identifierar återkommande mönster i barns meningsskapande: att utforska själv (en-skapar), att samspela med andra barn (kamrat-skapar) och att gemensamt skapa med pedagoger (barn-pedagog-skapar). Centrala strategier i dessa processer är upprepning, variation och syntes, vilka stödjer barns kroppsliga, kognitiva och begreppsliga utforskande av naturvetenskapliga fenomen såsom gravitation, friktion och magnetism. Studien visar även att pedagogernas roll är avgörande, när de fångar upp och utvecklar barns initiativ uppstår spontana och meningsfulla lärande-situationer, medan barns utforskande riskerar att förbises när pedagoger är frånvarande eller inte uppmärksammar barnens frågor.

    Studien bidrar till den växande forskningen om begynnande naturvetenskap genom att synliggöra hur de yngsta barnens vardagliga görande kan förstås som grundläggande naturvetenskapliga praktiker. De pedagogiska implikationerna pekar mot vikten av en lyhörd och närvarande pedagogik, tillgång till rika miljöer och material samt användningen av ett naturvetenskapligt språk för att successivt fördjupa barns begreppsliga förståelse. Studien framhåller förskolan som en central arena för att väcka nyfikenhet, förundran och långsiktigt intresse för naturvetenskap, inte genom att förmedla färdiga fakta, utan genom att skapa villkor för barns egna undersökande i dialog med andra barn, material och pedagoger.

    Ladda ner fulltext (pdf)
    fulltext
    Ladda ner (png)
    presentationsbild
  • Presentation: 2025-12-16 13:15 Ada Lovelace, B-building, LinköpingBeställ onlineKöp publikationen >>
    Oscar Colaco, Valency
    Linköpings universitet, Institutionen för datavetenskap, Programvara och system. Linköpings universitet, Tekniska fakulteten.
    Hardening Tree Ensembles: Real-Time and Effective Evasion Defences Beyond Adversarial Re-Training2025Licentiatavhandling, sammanläggning (Övrigt vetenskapligt)
    Abstract [en]

    Tree ensembles like random forests and gradient boosting machines are widely used machine learning (ML) models, often outperforming advanced techniques like deep neural networks on structured tabular data tasks. These models also have interpretable (human-understandable) structures that enable stakeholders to trace the decision-making process, making them particularly suitable for use in safety- and security-critical applications where trust in the model’s behaviour is paramount. Despite these advantages, recent work has shown that they are highly vulnerable to adversarial examples: carefully perturbed inputs that elicit misclassifications.

    These vulnerabilities are especially concerning as ML continues to permeate domains that are critical to societal functioning. Their seriousness is underscored by legislation such as the recently passed European Union Artificial Intelligence (AI) Act. This act mandates resilience against AI-specific vulnerabilities like evasion attacks caused by adversarial examples targeting ML models at inference time. Measures intended to improve resilience against such evasions, often referred to as hardening, generally involve two strategies: proactive defences, which aim to make models robust (e.g., adversarial re-training), and reactive defences, which focus on detecting and mitigating evasions at inference time. This thesis examines both strategies; it shows that proactive methods like model re-training are ineffective for tree ensembles and consequently advances the state-of-the-art in reactive defences.

    In the context of re-training, doubling the training set through targeted data augmentation steps left accuracy largely unchanged. However, robustness, when quantified using formal verification techniques, dropped by 28–82% across two case studies. This indicates that model re-training alone is ineffective for tree ensembles. To address this, we leveraged formal methods to develop Iceman, a prototype system that uses counterexample regions which violate the robustness property to detect evasion attempts. Iceman can detect evasion attacks regardless of the attack generation process without modifying the underlying tree ensemble. It outperforms the current state-of-the-art methods in evasion detection, OC-Score and GROOT. Across four case studies, it improves Matthews Correlation Coefficient scores by 0.20–0.91 and achieves detection speeds 5–115x faster than OC-Score. In addition, it provides alert filtering and prioritisation capabilities with over 98% accuracy to address alert fatigue in intrusion detection systems. However, Iceman’s applicability is limited to scenarios with fixed attacker perturbation budgets, characterised by pre-defined constraints on the input manipulations that an attacker can apply.

    To expand this applicability to unconstrained attacker perturbation budgets, we developed an additional system, called Maverick, designed to complement Iceman for a better defensive strategy. Just like Iceman, Maverick does not modify the underlying tree ensemble and can detect evasion attacks regardless of the attack generation process. We prove that Maverick’s core detection mechanism is mathematically equivalent to OC-Score, and present enhancements that achieve 85–563x speedups over OC-Score while maintaining identical detection performance and supporting evasion attack diagnostics with over 93% accuracy.

    Delarbeten
    1. Formal Verification of Tree Ensembles against Real-World Composite Geometric Perturbations
    Öppna denna publikation i ny flik eller fönster >>Formal Verification of Tree Ensembles against Real-World Composite Geometric Perturbations
    2023 (Engelska)Ingår i: Proceedings of the Workshop on Artificial Intelligence Safety 2023 (SafeAI 2023) co-located with the Thirty-Seventh AAAI Conference on Artificial Intelligence (AAAI 2023) / [ed] Pedroza G., Huang X., Chen X.C., Theodorou A., Hernandez-Orallo J., Castillo-Effen M., Mallah R., McDermid J., CEUR-WS , 2023, Vol. 3381, artikel-id 38Konferensbidrag, Publicerat paper (Refereegranskat)
    Abstract [en]

    Since machine learning components are now being considered for integration in safety-critical systems, safety stakeholdersshould be able to provide convincing arguments that the systems are safe for use in realistic deployment settings. In the caseof vision-based systems, the use of tree ensembles calls for formal stability verification against a host of composite geometricperturbations that the system may encounter. Such perturbations are a combination of an affine transformation like rotation,scaling, or translation and a pixel-wise transformation like changes in lighting. However, existing verification approachesmostly target small norm-based perturbations, and do not account for composite geometric perturbations. In this work,we present a novel method to precisely define the desired stability regions for these types of perturbations. We propose afeature space modelling process that generates abstract intervals which can be passed to VoTE, an efficient formal verificationengine that is specialised for tree ensembles. Our method is implemented as an extension to VoTE by defining a new propertychecker. The applicability of the method is demonstrated by verifying classifier stability and computing metrics associatedwith stability and correctness, i.e., robustness, fragility, vulnerability, and breakage, in two case studies. In both case studies,targeted data augmentation pre-processing steps were applied for robust model training. Our results show that even modelstrained with augmented data are unable to handle these types of perturbations, thereby emphasising the need for certifiedrobust training for tree ensembles.

    Ort, förlag, år, upplaga, sidor
    CEUR-WS, 2023
    Serie
    CEUR Workshop Proceedings, ISSN 1613-0073 ; 3381
    Nyckelord
    Machine Learning, Formal Verification, Tree Ensembles, Composite Perturbations, Geometric Perturbations, Random Forests, Gradient Boosting Machines, Semantic Perturbations, Stability, Robustness, Trustworthy AI, Trustworthy Computing
    Nationell ämneskategori
    Datavetenskap (datalogi)
    Identifikatorer
    urn:nbn:se:liu:diva-195996 (URN)2-s2.0-85159287306 (Scopus ID)
    Konferens
    The AAAI-23 Workshop on Artificial Intelligence Safety (SafeAI 2023), Washington DC, USA, February 13-14, 2023
    Forskningsfinansiär
    Wallenberg AI, Autonomous Systems and Software Program (WASP)
    Tillgänglig från: 2023-06-30 Skapad: 2023-06-30 Senast uppdaterad: 2025-11-13Bibliografiskt granskad
    2. Fast Evasion Detection & Alert Management in Tree-Ensemble-Based Intrusion Detection Systems
    Öppna denna publikation i ny flik eller fönster >>Fast Evasion Detection & Alert Management in Tree-Ensemble-Based Intrusion Detection Systems
    2024 (Engelska)Ingår i: 2024 IEEE 36TH INTERNATIONAL CONFERENCE ON TOOLS WITH ARTIFICIAL INTELLIGENCE, ICTAI, Institute of Electrical and Electronics Engineers (IEEE) , 2024, s. 404-412Konferensbidrag, Publicerat paper (Refereegranskat)
    Abstract [en]

    Intrusion Detection Systems (IDSs) can help bolster cyber resilience in high-risk systems by promptly detecting anomalies and thwarting security threats which could have catastrophic consequences. While Machine Learning (ML) techniques like Tree Ensembles are well suited for tasks like detecting anomalies, the widespread adoption of these techniques in IDSs faces barriers due to the threat of evasion attacks. Moreover, ML-based IDSs are susceptible to producing a high rate of false positive alerts during detection, causing alert fatigue. To alleviate these problems, we present a method that uses counterexample regions to detect evasion attacks in tree-ensemble-based IDSs. We generate these counterexample regions by defining a modified mapping checker in VoTE, a fast & scalable formal verification tool specialized for tree ensembles. Our method also provides quaternary annotations, empowering security managers with nuanced insights to better handle alerts in the triage queue. Our approach does not require training a separate model and displays good detection performance (≥98 %) in both adversarial & non-adversarial scenarios in four real-world case studies when compared to several approaches in the literature. The prototype system we implement based on our method called Iceman has a very low prediction latency, making it 5-115x faster than the current state-of-the-art in evasion detection for tree ensembles. Finally, empirical evaluations show that Iceman can correctly re-annotate the samples in the presence of evasion attacks for alert management purposes with an accuracy of more than 98 % .

    Ort, förlag, år, upplaga, sidor
    Institute of Electrical and Electronics Engineers (IEEE), 2024
    Serie
    Proceedings-International Conference on Tools With Artificial Intelligence, ISSN 1082-3409, E-ISSN 2375-0197
    Nyckelord
    Evasion Attacks; Adversarial Defences; Intrusion Detection Systems; Tree Ensembles; Formal Methods
    Nationell ämneskategori
    Datavetenskap (datalogi) Datorsystem
    Identifikatorer
    urn:nbn:se:liu:diva-211768 (URN)10.1109/ICTAI62512.2024.00065 (DOI)001447778900056 ()2-s2.0-85217421895 (Scopus ID)9798331527242 (ISBN)9798331527235 (ISBN)
    Konferens
    2024 IEEE 36th International Conference on Tools with Artificial Intelligence (ICTAI), Herndon, VA, OCT 28-30, 2024
    Forskningsfinansiär
    Wallenberg AI, Autonomous Systems and Software Program (WASP)
    Anmärkning

    Funding Agencies|Wallenberg AI, Autonomous Systems and Software Program (WASP) - Knut and Alice Wallenberg Foundation

    Tillgänglig från: 2025-02-20 Skapad: 2025-02-20 Senast uppdaterad: 2025-11-13
    3. Real-Time Evasion Detection in Tree Ensemble Automotive Intrusion Detection Systems
    Öppna denna publikation i ny flik eller fönster >>Real-Time Evasion Detection in Tree Ensemble Automotive Intrusion Detection Systems
    2025 (Engelska)Ingår i: 16th IEEE Vehicular Networking Conference (VNC), IEEE, 2025Konferensbidrag, Publicerat paper (Refereegranskat)
    Abstract [en]

    Safety-critical functions in modern vehicles rely on electronic control units that communicate using the controller area network (CAN) protocol, which lacks vital security features. In this context, machine learning (ML) based intrusion detection systems (IDSs) were proposed as a solution to improve cyber resilience through real-time attack detection. However, these ML-IDSs must also withstand evasion attacks that could compromise vehicular safety. To this end, this paper addresses such attacks in misuse-based tree ensemble IDSs and proposes a method that detects evasion attempts. It uses the ordered set of reached leaf nodes activated by correctly classified training samples as a normality baseline. An autoencoder-based detector then identifies deviations as likely evasion attempts. Our approach does not modify the protected tree ensemble IDS, assumes no knowledge of the process for generating adversarial examples (ensuring generalisability), and works with any additive tree ensemble. We also prove that it is mathematically equivalent to the state-of-the-art, which we advance in terms of detection speed by replacing its Hamming distance-based deviation search with an autoencoder-based model of typical predictive behavior trained using our custom loss function. This enhancement results in a detection process that is orders of magnitude faster. Additionally, our method offers nuanced insights regarding the pre-evasion attack signature prior to the adversarial perturbation, thereby enriching the security analysis of the features targeted during evasion attempts. The prototype system we present, called Maverick, has a very low prediction latency, making it 85-563x faster than the current state-of-the-art while maintaining identical detection accuracy. Finally, Maverick predicts the pre-evasion attack signatures of the evasion samples with an accuracy of more than 93% and has an average prediction time well below the message transmission rate for CAN 2.0 and CAN FD, thereby satisfying the criteria for an evasion-hardened & real-time automotive IDS.

    Ort, förlag, år, upplaga, sidor
    IEEE, 2025
    Serie
    IEEE Vehicular Networking Conference, ISSN 2157-9857, E-ISSN 2157-9865
    Nyckelord
    Tree Ensembles, Autoencoders, Intrusion Detection Systems, Real-time Systems, Safety, Security, Controller Area Networks, Adversarial Examples
    Nationell ämneskategori
    Datorsystem
    Identifikatorer
    urn:nbn:se:liu:diva-216350 (URN)10.1109/VNC64509.2025.11054177 (DOI)001540461700039 ()2-s2.0-105010777746 (Scopus ID)9798331524371 (ISBN)9798331524388 (ISBN)
    Konferens
    2025 IEEE Vehicular Networking Conference (VNC), Porto, Portugal, JUN 02-04, 2025
    Forskningsfinansiär
    Wallenberg AI, Autonomous Systems and Software Program (WASP)
    Anmärkning

    Funding Agencies|Wallenberg AI, Autonomous Systems and Software Program (WASP) - Knut and Alice Wallenberg Foundation

    Tillgänglig från: 2025-08-14 Skapad: 2025-08-14 Senast uppdaterad: 2025-11-13
    Ladda ner fulltext (pdf)
    fulltext
    Ladda ner (png)
    presentationsbild