Jun Hong's Outputs (25)

Leveraging activation and optimisation layers as dynamic strategies in the multi-task fuzzing scheme (2025)
Journal Article

Fuzzing is a common technique for identifying vulnerabilities in software. Recent approaches, like She et al.’s Multi-Task Fuzzing (MTFuzz), use neural networks to improve fuzzing efficiency. However, key elements like network architecture and hyperp... Read More about Leveraging activation and optimisation layers as dynamic strategies in the multi-task fuzzing scheme.

ProxiMix: Enhancing fairness with proximity samples in subgroups (2024)
Presentation / Conference Contribution

Many bias mitigation methods have been developed for addressing fairness issues in machine learning. We have found that using linear mixup alone, a data augmentation technique, for bias mitigation, can still retain biases present in dataset labels. R... Read More about ProxiMix: Enhancing fairness with proximity samples in subgroups.

Improving search space analysis of fuzzing mutators using cryptographic structures (2024)
Presentation / Conference Contribution

This paper introduces a novel approach to enhance the performance of software fuzzing mutator tools, by leveraging cryptographic structures known as substitution-permutation networks and Feistel networks. By integrating these structures into the exis... Read More about Improving search space analysis of fuzzing mutators using cryptographic structures.

AdaHAT: Adaptive hard attention to the task in task-incremental learning (2024)
Presentation / Conference Contribution

Catastrophic forgetting is a major issue in task-incremental learning, where a neural network loses what it has learned in previous tasks after being trained on new tasks. A number of architecture- based approaches have been proposed to address this... Read More about AdaHAT: Adaptive hard attention to the task in task-incremental learning.

Vulnerability detection through machine learning-based fuzzing: A systematic review (2024)
Journal Article

Modern software and networks underpin our digital society, yet the rapid growth of vulnerabilities that are uncovered within these threaten our cyber security posture. Addressing these issues at scale requires automated proactive approaches that can... Read More about Vulnerability detection through machine learning-based fuzzing: A systematic review.

What will make misinformation spread: An XAI perspective (2023)
Presentation / Conference Contribution

Explainable Artificial Intelligence (XAI) techniques can provide explanations of how AI systems or models make decisions, or what factors AI considers when making the decisions. Online social networks have a problem with misinformation which is known... Read More about What will make misinformation spread: An XAI perspective.

Ego-graph replay based continual learning for misinformation engagement prediction (2022)
Presentation / Conference Contribution

Online social network platforms have a problem with misinformation. One popular way of addressing this problem is via the use of machine learning based automated misinformation detection systems to classify if a post is misinformation. Instead of pos... Read More about Ego-graph replay based continual learning for misinformation engagement prediction.

Ego-graph replay based continual learning for misinformation engagement prediction (2022)
Presentation / Conference Contribution

Online social network platforms have a problem with misinformation. One popular way of addressing this problem is via the use of machine learning based automated misinfor-mation detection systems to classify if a post is misinformation. Instead of po... Read More about Ego-graph replay based continual learning for misinformation engagement prediction.

Social influence prediction with train and test time augmentation for graph neural networks (2021)
Presentation / Conference Contribution

Data augmentation has been widely used in machine learning for natural language processing and computer vision tasks to improve model performance. However, little research has studied data augmentation on graph neural networks, particularly using aug... Read More about Social influence prediction with train and test time augmentation for graph neural networks.

Towards establishing a 'cooperation' measure for coupled movement in close-proximity human-robot interaction (2020)
Presentation / Conference Contribution

To achieve safe close-proximity human-robot interaction , particularly for physically assitive tasks, robot motion planning needs to recognize and adapt to the behaviour of humans in the long-term. Generally, motion prediction with probabilistic conf... Read More about Towards establishing a 'cooperation' measure for coupled movement in close-proximity human-robot interaction.

Social network influence ranking via embedding network interactions for user recommendation (2020)
Presentation / Conference Contribution

Within social networks user influence may be modelled based on user interactions. Further, it is typical to recommend users to others. What is the role of user influence in user recommendation In this paper, we first propose to use a node embedding a... Read More about Social network influence ranking via embedding network interactions for user recommendation.

Privacy preserving record linkage in the presence of missing values (2017)
Journal Article

© 2017 The problem of record linkage is to identify records from two datasets, which refer to the same entities (e.g. patients). A particular issue of record linkage is the presence of missing values in records, which has not been fully addressed. An... Read More about Privacy preserving record linkage in the presence of missing values.

A novel ensemble learning approach to unsupervised record linkage (2017)
Journal Article

© 2017 Record linkage is a process of identifying records that refer to the same real-world entity. Many existing approaches to record linkage apply supervised machine learning techniques to generate a classification model that classifies a pair of r... Read More about A novel ensemble learning approach to unsupervised record linkage.

The event calculus in probabilistic logic programming with annotated disjunctions (2017)
Presentation / Conference Contribution

We propose a new probabilistic extension to the event calculus using the probabilistic logic programming (PLP) language ProbLog, and a language construct called the annotated disjunction. This is the first extension of the event calculus capable of h... Read More about The event calculus in probabilistic logic programming with annotated disjunctions.

Managing different sources of uncertainty in a BDI framework in a principled way with tractable fragments (2017)
Journal Article

The Belief-Desire-Intention (BDI) architecture is a practical approach for modelling large-scale intelligent systems. In the BDI setting, a complex system is represented as a network of interacting agents – or components – each one modelled based on... Read More about Managing different sources of uncertainty in a BDI framework in a principled way with tractable fragments.

Tweet for behavior change: Using social media for the dissemination of public health messages (2017)
Journal Article

Background: Social media public health campaigns have the advantage of tailored messaging at low cost and large reach, but little is known about what would determine their feasibility as tools for inducing attitude and behavior change.
Objective: Th... Read More about Tweet for behavior change: Using social media for the dissemination of public health messages.

A collaborative multiagent framework based on online risk-aware planning and decision-making (2017)
Journal Article

Planning is an essential process in teams of multiple agents pursuing a common goal. When the effects of actions undertaken by agents are uncertain, evaluating the potential risk of such actions alongside their utility might lead to more rational dec... Read More about A collaborative multiagent framework based on online risk-aware planning and decision-making.

Context-dependent combination of sensor information in Dempster–Shafer theory for BDI (2016)
Journal Article

© 2016, The Author(s). There has been much interest in the belief–desire–intention (BDI) agent-based model for developing scalable intelligent systems, e.g. using the AgentSpeak framework. However, reasoning from sensor information in these large-sca... Read More about Context-dependent combination of sensor information in Dempster–Shafer theory for BDI.

Uncertain information combination for decision making in smart grid BDI agent systems (2016)
Journal Article

In a smart grid SCADA (supervisory control and data acquisition) system, sensor information (e.g. temperature, voltage, frequency, etc.) from heterogeneous sources can be used to reason about the true system state (e.g. faults, attacks, etc.). Before... Read More about Uncertain information combination for decision making in smart grid BDI agent systems.

Probabilistic planning in agentspeak using the POMDP framework (2016)
Presentation / Conference Contribution

© Springer International Publishing Switzerland 2016. AgentSpeak is a logic-based programming language, based on the Belief- Desire-Intention paradigm, suitable for building complex agent-based systems. To limit the computational complexity, agents i... Read More about Probabilistic planning in agentspeak using the POMDP framework.