Christopher Leckie

Christopher Leckie
University of Melbourne | MSD · Department of Computing and Information Systems

About

403
Publications
111,906
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
13,315
Citations

Publications

Publications (403)
Article
The emergence of social media as one of the main platforms for people to access news has enabled the wide dissemination of fake news, having serious impacts on society. Thus, it is really important to identify fake news with high confidence in a timely manner, which is not feasible using manual analysis. This has motivated numerous studies on autom...
Conference Paper
Full-text available
Voltage calculations are key for most distribution network analyses. However, the challenge in low voltage (LV) networks is that electrical models are not readily available and, therefore, accurate voltage calculations (via power flows) are not possible. Alternatively, regression methods can be used to capture the relationships among the historical...
Conference Paper
Full-text available
The adoption of distributed energy resources creates opportunities for end users to provide system-level services through aggregators. Nonetheless, simultaneous power exports/imports from end users can cause challenges in low voltage (LV) networks. While operating envelopes (OEs), i.e., time-varying export/import limits, can help, their correct cal...
Chapter
Clustering is a process of finding groups of similar objects in a given dataset. Finding clusters in graphs, especially crisp clusters, which have minimal or no overlapping clusters, is challenging. Further, clustering is an ill-defined problem, resulting in multiple possible solutions for the same dataset. Hence, a challenge here is that the possi...
Presentation
Full-text available
The increasing adoption of residential rooftop solar PV and batteries in Australia has led to a paradigm shift in how distribution companies (known as Distribution Network Service Providers [DNSPs]) will set export limits. Instead of using static (fixed) export limits designed to ensure the integrity of the poles and wires for a few critical hours...
Conference Paper
Full-text available
The proliferation of residential distributed energy resources (DER), such as rooftop solar PV, batteries, and electric vehicles (EVs), is creating significant challenges for distribution companies. This is because the electricity distribution networks (i.e., the poles and wires connecting homes and businesses) have not been designed for the diverse...
Article
Full-text available
It has been shown that neural networks are vulnerable to adversarial attacks: adding well-crafted, imperceptible perturbations to their input can modify their output. Adversarial training is one of the most effective approaches to training robust models against such attacks. Unfortunately, this method is much slower than vanilla training of neural...
Technical Report
Full-text available
This is the final report of the "Model-Free Operating Envelopes at NMI Level" Project funded by C4NET. The report presents the final version of the model-free operating envelope (OE) approach developed throughout this project and presents the main findings of the last 6 months of this project.
Article
Cross-domain graph anomaly detection (CD-GAD) describes the problem of detecting anomalous nodes in an unlabelled target graph using auxiliary, related source graphs with labelled anomalous and normal nodes. Although it presents a promising approach to address the notoriously high false positive issue in anomaly detection, little work has been done...
Article
Full-text available
Machine learning (ML) methods such as artificial neural networks are rapidly becoming ubiquitous in modern science, technology, and industry. Despite their accuracy and sophistication, neural networks can be easily fooled by carefully designed malicious inputs known as adversarial attacks. While such vulnerabilities remain a serious challenge for c...
Preprint
Full-text available
Machine learning algorithms are powerful tools for data driven tasks such as image classification and feature detection, however their vulnerability to adversarial examples - input samples manipulated to fool the algorithm - remains a serious challenge. The integration of machine learning with quantum computing has the potential to yield tools offe...
Preprint
Full-text available
The emergence of social media as one of the main platforms for people to access news has enabled the wide dissemination of fake news. This has motivated numerous studies on automating fake news detection. Although there have been limited attempts at unsupervised fake news detection, their performance suffers due to not exploiting the knowledge from...
Preprint
Full-text available
The analysis of distributed techniques is often focused upon their efficiency, without considering their robustness (or lack thereof). Such a consideration is particularly important when devices or central servers can fail, which can potentially cripple distributed systems. When such failures arise in wireless communications networks, important ser...
Chapter
Deep neural network (DNN) classifiers are vulnerable to backdoor attacks. An adversary poisons some of the training data in such attacks by installing a trigger. The goal is to make the trained DNN output the attacker’s desired class whenever the trigger is activated while performing as usual for clean data. Various approaches have recently been pr...
Presentation
Full-text available
The project “Model-Free Operating Envelopes at NMI Level” has demonstrated it is possible to capture the physics of three-phase LV networks from smart meter data using neural networks to create a fast and accurate electrical model-free approach. This enables us to calculate voltages allowing the calculation of DER hosting capacity and operating env...
Technical Report
Full-text available
This report builds on the previous two reports (foundations, methodology, and extensive performance tests) and presents improvements and updates on the Project “Model-Free Operating Envelopes at NMI Level”. Specifically, this report presents improvements to the offline data pipeline previously defined, presents an alternative allocation technique t...
Chapter
Securing communication networks has become increasingly important due to the growth in cybersecurity attacks, such as ransomware and denial of service attacks. In order to better observe, detect and track attacks in large networks, accurate and efficient anomaly detection algorithms are needed. In this paper, we address how the redundancy of the no...
Article
Full-text available
The proliferation of residential technologies such as photovoltaic (PV) systems and electric vehicles can cause voltage issues in low voltage (LV) networks. During operation, voltage calculations can help determining settings, such as PV curtailment, that ensure compliance with statutory limits. In planning, voltage calculations can help assessing...
Preprint
Full-text available
Cross-domain graph anomaly detection (CD-GAD) describes the problem of detecting anomalous nodes in an unlabelled target graph using auxiliary, related source graphs with labelled anomalous and normal nodes. Although it presents a promising approach to address the notoriously high false positive issue in anomaly detection, little work has been done...
Preprint
Full-text available
Machine learning (ML) methods such as artificial neural networks are rapidly becoming ubiquitous in modern science, technology and industry. Despite their accuracy and sophistication, neural networks can be easily fooled by carefully designed malicious inputs known as adversarial attacks. While such vulnerabilities remain a serious challenge for cl...
Chapter
Neural networks are vulnerable to adversarial attacks: adding well-crafted, imperceptible perturbations to their input can modify their output. Adversarial training is one of the most effective approaches in training robust models against such attacks. However, it is much slower than vanilla training of neural networks since it needs to construct a...
Preprint
Deep neural network (DNN) classifiers are vulnerable to backdoor attacks. An adversary poisons some of the training data in such attacks by installing a trigger. The goal is to make the trained DNN output the attacker's desired class whenever the trigger is activated while performing as usual for clean data. Various approaches have recently been pr...
Preprint
Neural networks are vulnerable to adversarial attacks: adding well-crafted, imperceptible perturbations to their input can modify their output. Adversarial training is one of the most effective approaches to training robust models against such attacks. Unfortunately, this method is much slower than vanilla training of neural networks since it needs...
Technical Report
Full-text available
This report presents the different technical aspects and results associated with the production of neural networks used in the electrical model-free approach to calculate voltages proposed by the University of Melbourne within the context of the Project “Model-Free Operating Envelopes at NMI Level”. The report investigates single LV circuits and it...
Article
The predict+optimize problem combines machine learning and combinatorial optimization by predicting the problem coefficients first and then using these coefficients to solve the optimization problem. While this problem can be solved in two separate stages, recent research shows end to end models can achieve better results. This requires differentia...
Article
Full-text available
Deep Neural Networks (DNNs) generally require large-scale datasets for training. Since manually obtaining clean labels for large datasets is extremely expensive, unsupervised models based on domain-specific heuristics can be used to efficiently infer the labels for such datasets. However, the labels from such inferred sources are typically noisy, w...
Chapter
Blockmodelling is the process of determining community structure in a graph. Real graphs contain noise and so it is up to the blockmodelling method to allow for this noise and reconstruct the most likely role memberships and role relationships. Relationships are encoded in a graph using the absence and presence of edges. Two objects are considered...
Presentation
Full-text available
Distribution companies, who manage the poles and wires, struggle to have accurate and up-to-date electrical models of their residential areas, known as low voltage (LV) networks. And without electrical models, it is hard to assess the hosting capacity for distributed energy resources (DER) such as solar PV or electric vehicles; particularly when vo...
Chapter
Recent unsupervised GNN based graph anomaly detection (GAD) methods adopt specific mechanisms designed for anomaly detection. This is in contrast to earlier methods that utilise components such as graph autoencoders that were designed for more general use-cases. However, these newer methods only lead to a modest increase in detection accuracy at th...
Article
An important factor that discourages patrons from using bus systems is the long and uncertain waiting times. Therefore, accurate bus travel time prediction is important to improve the serviceability of bus transport systems. Many researchers have proposed machine learning and deep learning-based models for bus travel time predictions. However, most...
Technical Report
Full-text available
This report provides the foundations of the electrical model-free approach to calculate voltages proposed by the University of Melbourne within the context of the Project “Model-Free Operating Envelopes at NMI Level”. Furthermore, this report presents the analyses of the data received from the Victorian Distribution Network Service Providers (DNSPs...
Preprint
Neural networks are vulnerable to adversarial attacks: adding well-crafted, imperceptible perturbations to their input can modify their output. Adversarial training is one of the most effective approaches in training robust models against such attacks. However, it is much slower than vanilla training of neural networks since it needs to construct a...
Chapter
With the advent of 5th Generation (5G) of mobile networks, a diverse range of new computer networking technologies are being devised to meet the stringent demands of applications that require ultra-low latency, high bandwidth and geolocation-based services. Mobile Edge Computing (MEC) is a prominent example of such an emerging technology, which pro...
Article
Generative Adversarial Networks (GANs) have seen great research interest in recent years, due to both their ability to represent structure in data and generate novel samples. Anomaly detection, which discerns novel samples or patterns, is a well-known problem that can be studied using GANs with a fresh perspective, especially in novel application d...
Preprint
The vulnerability of machine learning models to adversarial perturbations has motivated a significant amount of research under the broad umbrella of adversarial machine learning. Sophisticated attacks may cause learning algorithms to learn decision functions or make decisions with poor predictive performance. In this context, there is a growing bod...
Chapter
The prevalence of fake news over social media has a profound impact on justice, public trust and society as a whole. Although significant effort has been applied to mitigate its negative impact, our study shows that existing fake news detection algorithms may perform poorly on new data. In other words, the performance of a model trained on one data...
Article
Many recent studies have demonstrated that the propagation patterns of news on social media can facilitate the detection of fake news. Most of these studies rely on the complete propagation networks to build their model, which is not fully available in the early stages and may take a long time to complete. Hence, relying on the complete propagation...
Article
Shape analogy is a key technique in analyzing time series. That is, time series are compared by how much they look alike. This concept has been applied for many years in geometry. Notably, none of the current techniques describe a time series as a geometric curve that is expressed by its relative location and form in space. To fill this gap, we int...
Preprint
Full-text available
Recent years have witnessed the significant damage caused by various types of fake news. Although considerable effort has been applied to address this issue and much progress has been made on detecting fake news, most existing approaches mainly rely on the textual content and/or social context, while knowledge-level information---entities extracted...
Conference Paper
Nonlinear regression, although widely used in engineering, financial and security applications for automated decision making, is known to be vulnerable to training data poisoning. Targeted poisoning attacks may cause learning algorithms to fit decision functions with poor predictive performance. This paper presents a new analysis of local intrinsic...
Article
Full-text available
With the rapid evolution of social media, fake news has become a significant social problem, which cannot be addressed in a timely manner using manual investigation. This has motivated numerous studies on automating fake news detection. Most studies explore supervised training models with different modalities (e.g., text, images, and propagation ne...
Chapter
Market Basket Analysis (MBA) is a popular technique to identify associations between products, which is crucial for business decision making. Previous studies typically adopt conventional frequent itemset mining algorithms to perform MBA. However, they generally fail to uncover rarely occurring associations among the products at their most granular...
Article
Support Vector Machines (SVMs) are vulnerable to targeted training data manipulations such as poisoning attacks and label flips. By carefully manipulating a subset of training samples, the attacker forces the learner to compute an incorrect decision boundary, thereby causing misclassifications. Considering the increased importance of SVMs in engine...
Preprint
Full-text available
With the rapid evolution of social media, fake news has become a significant social problem, which cannot be addressed in a timely manner using manual investigation. This has motivated numerous studies on automating fake news detection. Most studies explore supervised training models with different modalities (e.g., text, images, and propagation ne...
Chapter
Cognitive radio networks can be used to detect anomalous and adversarial communications to achieve situational awareness on the radio frequency spectrum. This paper proposes a distributed anomaly detection scheme based on adversarially-trained data models. While many anomaly detection methods typically depend on a central decision-making server, ou...
Preprint
Full-text available
The predict+optimize problem combines machine learning ofproblem coefficients with a combinatorial optimization prob-lem that uses the predicted coefficients. While this problemcan be solved in two separate stages, it is better to directlyminimize the optimization loss. However, this requires dif-ferentiating through a discrete, non-differentiable...
Preprint
Contrast pattern mining (CPM) aims to discover patterns whose support increases significantly from a background dataset compared to a target dataset. CPM is particularly useful for characterising changes in evolving systems, e.g., in network traffic analysis to detect unusual activity. While most existing techniques focus on extracting either the w...
Article
Short bursts of repeating patterns [intervals of recurrence (IoR)] manifest themselves in many applications, such as in the time-series data captured from an athlete’s movements using a wearable sensor while performing exercises. We present an efficient, online, one-pass, and real-time algorithm for finding and tracking IoR in a time-series data st...
Preprint
Regression models, which are widely used from engineering applications to financial forecasting, are vulnerable to targeted malicious attacks such as training data poisoning, through which adversaries can manipulate their predictions. Previous works that attempt to address this problem rely on assumptions about the nature of the attack/attacker or...
Preprint
Support Vector Machines (SVMs) are vulnerable to targeted training data manipulations such as poisoning attacks and label flips. By carefully manipulating a subset of training samples, the attacker forces the learner to compute an incorrect decision boundary, thereby cause misclassifications. Considering the increased importance of SVMs in engineer...
Preprint
Decision tree learning is a widely used approach in machine learning, favoured in applications that require concise and interpretable models. Heuristic methods are traditionally used to quickly produce models with reasonably high accuracy. A commonly criticised point, however, is that the resulting trees may not necessarily be the best representati...
Preprint
Full-text available
Many learning tasks involve multi-modal data streams, where continuous data from different modes convey a comprehensive description about objects. A major challenge in this context is how to efficiently interpret multi-modal information in complex environments. This has motivated numerous studies on learning unsupervised representations from multi-...
Conference Paper
With the rapid adoption of next generation networking architectures in 5G, like Multi-Access Edge Computing (MEC), there is a shift in the core processing capabilities to the edge of the network. This helps facilitate higher bandwidth and ultra-low latency responses, but can increase the attack surface for cyber-attacks like Denial of Service (DoS)...
Preprint
Deep learning classifiers are susceptible to well-crafted, imperceptible variations of their inputs, known as adversarial attacks. In this regard, the study of powerful attack models sheds light on the sources of vulnerability in these classifiers, hopefully leading to more robust ones. In this paper, we introduce AdvFlow: a novel black-box adversa...
Preprint
Although significant effort has been applied to fact-checking, the prevalence of fake news over social media, which has profound impact on justice, public trust and our society as a whole, remains a serious problem. In this work, we focus on propagation-based fake news detection, as recent studies have demonstrated that fake news and real news spre...
Preprint
Deep neural network classifiers suffer from adversarial vulnerability: well-crafted, unnoticeable changes to the input data can affect the classifier decision. In this regard, the study of powerful adversarial attacks can help shed light on sources of this malicious behavior. In this paper, we propose a novel black-box adversarial attack using norm...
Preprint
Full-text available
Market Basket Analysis (MBA) is a popular technique to identify associations between products, which is crucial for business decision making. Previous studies typically adopt conventional frequent itemset mining algorithms to perform MBA. However, they generally fail to uncover rarely occurring associations among the products at their most granular...
Chapter
Events detected from social media streams often include early signs of accidents, crimes or disasters. Therefore, they can be used by related parties for timely and efficient response. Although significant progress has been made on event detection from tweet streams, most existing methods have not considered the posted images in tweets, which provi...
Article
We study the predict+optimise problem, where machine learning and combinatorial optimisation must interact to achieve a common goal. These problems are important when optimisation needs to be performed on input parameters that are not fully observed but must instead be estimated using machine learning. We provide a novel learning technique for pred...
Chapter
Full-text available
Unsupervised anomaly detection is commonly performed using a distance or density based technique, such as K-Nearest neighbours, Local Outlier Factor or One-class Support Vector Machines. One-class Support Vector Machines reduce the computational cost of testing new data by providing sparse solutions. However, all these techniques have relatively hi...
Article
Full-text available
Users need to understand the predictions of a classifier, especially when decisions based on the predictions can have severe consequences. The explanation of a prediction reveals the reason why a classifier makes a certain prediction, and it helps users to accept or reject the prediction with greater confidence. This paper proposes an explanation m...
Preprint
Events detected from social media streams often include early signs of accidents, crimes or disasters. Therefore, they can be used by related parties for timely and efficient response. Although significant progress has been made on event detection from tweet streams, most existing methods have not considered the posted images in tweets, which provi...
Article
Full-text available
A critical problem in time series analysis is change point detection, which identifies the times when the underlying distribution of a time series abruptly changes. However, several shortcomings limit the use of some existing techniques in real-world applications. First, several change point detection techniques are offline methods, where the whole...
Preprint
Normalizing flows attempt to model an arbitrary probability distribution through a set of invertible mappings. These transformations are required to achieve a tractable Jacobian determinant that can be used in high-dimensional scenarios. The first normalizing flow designs used coupling layer mappings built upon affine transformations. The significa...
Chapter
Coresets are representative samples of data that can be used to train machine learning models with provable guarantees of approximating the accuracy of training on the full data set. They have been used for scalable clustering of large datasets and result in better cluster partitions compared to clustering a random sample. In this paper, we present...
Conference Paper
Deployment and management of large-scale mobile edge computing infrastructure in 5G networks has created a major challenge for mobile operators. The ability to extract common users' trajectories (i.e., corridors) in mobile networks helps mobile operators to better manage and orchestrate the allocation of network resources. However, compared with ot...