Hussein Abbass

Duy Tung Nguyen, Kathryn Kasmarik, Hussein Abbass

Neural-based learning agents make decisions using internal artificial neural networks. In certain situations, it becomes pertinent that this knowledge is re-interpreted in a friendly form to both the human and the machine. These situations include: when agents are required to communicate the knowledge they learn to each other in a transparent way in the presence of an external human observer, in human-machine teaming settings where humans and machines need to collaborate on a task, or where there is a requirement to verify the knowledge exchanged between the agents. We propose an interpretable knowledge fusion framework suited for neural-based learning agents, and propose a Priority on Weak State Areas (PoWSA) retraining technique. We first test the proposed framework on a synthetic binary classification task before evaluating it on a shepherding-based multi-agent swarm guidance task. Results demonstrate that the proposed framework increases the success rate on the swarm-guidance environment by 11% and better stability in return for a modest increase in computational cost of 14.5% to achieve interpretability. Moreover, the framework presents the knowledge learnt by an agent in a human-friendly representation, leading to a better descriptive visual representation of an agent's knowledge.

Adam Hepworth, Aya Hussein, Darryn Reid et al.

Contemporary swarm indicators are often used in isolation, focused on extracting information at the individual or collective levels. Consequently, these are seldom integrated to infer a top-level operating picture of the swarm, its members, and its overall collective dynamics. The primary contribution of this paper is to organise a suite of indicators about swarms into an ontologically-arranged collection of information markers to characterise the swarm from the perspective of an external observer\textemdash, a recognition agent. Our contribution shows the foundations for a new area of research that we tile swarm analytics, whose primary concern is with the design and organisation of collections of swarm markers to understand, detect, recognise, track, and learn a particular insight about a swarm system. We present our designed framework of information markers that offer a new avenue for swarm research, especially for heterogeneous and cognitive swarms that may require more advanced capabilities to detect agencies and categorise agent influences and responses.

Adam Hepworth, Aya Hussein, Darryn Reid et al.

An emerging challenge in swarm shepherding research is to design effective and efficient artificial intelligence algorithms that maintain a low-computational ceiling while increasing the swarm's abilities to operate in diverse contexts. We propose a methodology to design a context-aware swarm-control intelligent agent. The intelligent control agent (shepherd) first uses swarm metrics to recognise the type of swarm it interacts with to then select a suitable parameterisation from its behavioural library for that particular swarm type. The design principle of our methodology is to increase the situation awareness (i.e. information contents) of the control agent without sacrificing the low-computational cost necessary for efficient swarm control. We demonstrate successful shepherding in both homogeneous and heterogeneous swarms.

Hassan Wasswa, Hussein Abbass, Timothy Lynar

In an effort to counter the increasing IoT botnet-based attacks, state-of-the-art deep learning methods have been proposed and have achieved impressive detection accuracy. However, their computational intensity restricts deployment on resource-constrained IoT devices, creating a critical need for lightweight detection models. A common solution to this challenge is model compression via quantization. This study proposes a VAE-MLP model framework where an MLP-based classifier is trained on 8-dimensional latent vectors derived from the high-dimensional train data using the encoder component of a pretrained variational autoencoder (VAE). Two widely used quantization strategies--Quantization-Aware Training (QAT) and Post-Training Quantization (PTQ)--are then systematically evaluated in terms of their impact on detection performance, storage efficiency, and inference latency using two benchmark IoT botnet datasets--N-BaIoT and CICIoT2022. The results revealed that, with respect to detection accuracy, the QAT strategy experienced a more noticeable decline,whereas PTQ incurred only a marginal reduction compared to the original unquantized model. Furthermore, PTQ yielded a 6x speedup and 21x reduction in size, while QAT achieved a 3x speedup and 24x compression, demonstrating the practicality of quantization for device-level IoT botnet detection.

Adam Hepworth, Zena Assaad, Austin Wyatt et al.

Military human robot interaction (MHRI) presents a novel opportunity to blend the capabilities of autonomous and Artificial Intelligence (AI)-enabled systems with the skills and expertise of humans. The concept promises military advantages and greater operational effectiveness and efficiencies. However, the associated human-AI dynamics create challenges when attempting to design, implement, and operationalise the increasingly symbiotic relationship between humans and machines. Meaningful human control (MHC) is a popularised conceptualisation of what is deemed a responsible interaction among human and artificial agents; however, this notion falls short in military contexts and hinders the realisation of military advantages that could be achieved by advancing the adoption of responsible AI. This paper presents meaningful human command (MHC1) as a more operationally effective concept for advanced military command and control systems that embed AI-enabled autonomous systems. We introduce, explore, and unpack meaningful human command in the context of military human-robot interaction, presenting a vignette that offers a technologically feasible concept of an AI-enabled system within military operations. The vignette is used to guide, contextualise, and add realism to the narrative describing the concept and highlights associated MHRI challenges.

Hassan Wasswa, Timothy Lynar, Hussein Abbass

The Internet of Things (IoT) technology has rapidly gained popularity with applications widespread across a variety of industries. However, IoT devices have been recently serving as a porous layer for many malicious attacks to both personal and enterprise information systems with the most famous attacks being botnet-related attacks. The work in this study leveraged Variational Auto-encoder (VAE) and cost-sensitive learning to develop lightweight, yet effective, models for IoT-botnet detection. The aim is to enhance the detection of minority class attack traffic instances which are often missed by machine learning models. The proposed approach is evaluated on a multi-class problem setting for the detection of traffic categories on highly imbalanced datasets. The performance of two deep learning models including the standard feed forward deep neural network (DNN), and Bidirectional-LSTM (BLSTM) was evaluated and both recorded commendable results in terms of accuracy, precision, recall and F1-score for all traffic classes.

Hassan Wasswa, Timothy Lynar, Aziida Nanyonga et al.

Despite the demonstrated effectiveness of transformer models in NLP, and image and video classification, the available tools for extracting features from captured IoT network flow packets fail to capture sequential patterns in addition to the absence of spatial patterns consequently limiting transformer model application. This work introduces a novel preprocessing method to adapt transformer models, the vision transformer (ViT) in particular, for IoT botnet attack detection using network flow packets. The approach involves feature extraction from .pcap files and transforming each instance into a 1-channel 2D image shape, enabling ViT-based classification. Also, the ViT model was enhanced to allow use any classifier besides Multilayer Perceptron (MLP) that was deployed in the initial ViT paper. Models including the conventional feed forward Deep Neural Network (DNN), LSTM and Bidirectional-LSTM (BLSTM) demonstrated competitive performance in terms of precision, recall, and F1-score for multiclass-based attack detection when evaluated on two IoT attack datasets.

Hassan Wasswa, Hussein Abbass, Timothy Lynar

Due to the exponential rise in IoT-based botnet attacks, researchers have explored various advanced techniques for both dimensionality reduction and attack detection to enhance IoT security. Among these, Variational Autoencoders (VAE), Vision Transformers (ViT), and Graph Neural Networks (GNN), including Graph Convolutional Networks (GCN) and Graph Attention Networks (GAT), have garnered significant research attention in the domain of attack detection. This study evaluates the effectiveness of four state-of-the-art deep learning architectures for IoT botnet detection: a VAE encoder with a Multi-Layer Perceptron (MLP), a VAE encoder with a GCN, a VAE encoder with a GAT, and a ViT encoder with an MLP. The evaluation is conducted on a widely studied IoT benchmark dataset--the N-BaIoT dataset for both binary and multiclass tasks. For the binary classification task, all models achieved over 99.93% in accuracy, recall, precision, and F1-score, with no notable differences in performance. In contrast, for the multiclass classification task, GNN-based models showed significantly lower performance compared to VAE-MLP and ViT-MLP, with accuracies of 86.42%, 89.46%, 99.72%, and 98.38% for VAE-GCN, VAE-GAT, VAE-MLP, and ViT-MLP, respectively.

Hassan Wasswa, Hussein Abbass, Timothy Lynar

With the rise of IoT-based botnet attacks, researchers have explored various learning models for detection, including traditional machine learning, deep learning, and hybrid approaches. A key advancement involves deploying attention mechanisms to capture long-term dependencies among features, significantly improving detection accuracy. However, most models treat attack instances independently, overlooking inter-instance relationships. Graph Neural Networks (GNNs) address this limitation by learning an embedding space via iterative message passing where similar instances are placed closer based on node features and relationships, enhancing classification performance. To further improve detection, attention mechanisms have been embedded within GNNs, leveraging both long-range dependencies and inter-instance connections. However, transforming the high dimensional IoT attack datasets into a graph structured dataset poses challenges, such as large graph structures leading computational overhead. To mitigate this, this paper proposes a framework that first reduces dimensionality of the NetFlow-based IoT attack dataset before transforming it into a graph dataset. We evaluate three dimension reduction techniques--Variational Autoencoder (VAE-encoder), classical autoencoder (AE-encoder), and Principal Component Analysis (PCA)--and compare their effects on a Graph Attention neural network (GAT) model for botnet attack detection

Taylan Akay, Harrison Tolley, Hussein Abbass

In the age of AI, human commanders need to use the computational powers available in today's environment to simulate a very large number of scenarios. Within each scenario, situations occur where different decision design options could have ethical consequences. Making these decisions reliant on human judgement is both counter-productive to the aim of exploring very large number of scenarios in a timely manner and infeasible when considering the workload needed to involve humans in each of these choices. In this paper, we move human judgement outside the simulation decision cycle. Basically, the human will design the ethical metric space, leaving it to the simulated environment to explore the space. When the simulation completes its testing cycles, the testing environment will come back to the human commander with a few options to select from. The human commander will then exercise human-judgement to select the most appropriate course of action, which will then get executed accordingly. We assume that the problem of designing metrics that are sufficiently granular to assess the ethical implications of decisions is solved. Subsequently, the fundamental problem we look at in this paper is how to weight ethical decisions during the running of these simulations; that is, how to dynamically weight the ethical attributes when agents are faced with decision options with ethical implications during generative simulations. The multi-criteria decision making literature has started to look at nearby problems, where the concept of entropy has been used to determine the weights during aggregation. We draw from that literature different approaches to automatically calculate the weights for ethical attributes during simulation-based testing and evaluation.

Jacob Simpson, Rudolph Oosthuizen, Sondoss El Sawah et al.

Artificial Intelligence (AI) is rapidly becoming integrated into military Command and Control (C2) systems as a strategic priority for many defence forces. The successful implementation of AI is promising to herald a significant leap in C2 agility through automation. However, realistic expectations need to be set on what AI can achieve in the foreseeable future. This paper will argue that AI could lead to a fragility trap, whereby the delegation of C2 functions to an AI could increase the fragility of C2, resulting in catastrophic strategic failures. This calls for a new framework for AI in C2 to avoid this trap. We will argue that antifragility along with agility should form the core design principles for AI-enabled C2 systems. This duality is termed Agile, Antifragile, AI-Enabled Command and Control (A3IC2). An A3IC2 system continuously improves its capacity to perform in the face of shocks and surprises through overcompensation from feedback during the C2 decision-making cycle. An A3IC2 system will not only be able to survive within a complex operational environment, it will also thrive, benefiting from the inevitable shocks and volatility of war.

Hung The Nguyen, Matthew Garratt, Lam Thu Bui et al.

The guidance of a large swarm is a challenging control problem. Shepherding offers one approach to guide a large swarm using a few shepherding agents (sheepdogs). While noise is an inherent characteristic in many real-world problems, the impact of noise on shepherding is not a well-studied problem. We study two forms of noise. First, we evaluate noise in the sensorial information received by the shepherd about the location of sheep. Second, we evaluate noise in the ability of the sheepdog to influence sheep due to disturbance forces occurring during actuation. We study both types of noise in this paper, and investigate the performance of Strömbom's approach under these actuation and perception noises. To ensure that the parameterisation of the algorithm creates a stable performance, we need to run a large number of simulations, while increasing the number of random episodes until stability is achieved. We then systematically study the impact of sensorial and actuation noise on performance. Strömbom's approach is found to be more sensitive to actuation noise than perception noise. This implies that it is more important for the shepherding agent to influence the sheep more accurately by reducing actuation noise than attempting to reduce noise in its sensors. Moreover, different levels of noise required different parameterisation for the shepherding agent, where the threshold needed by an agent to decide whether or not to collect astray sheep is different for different noise levels.

Saber Elsayed, Hemant Singh, Essam Debie et al.

Shepherding involves herding a swarm of agents (\emph{sheep}) by another a control agent (\emph{sheepdog}) towards a goal. Multiple approaches have been documented in the literature to model this behaviour. In this paper, we present a modification to a well-known shepherding approach, and show, via simulation, that this modification improves shepherding efficacy. We then argue that given complexity arising from obstacles laden environments, path planning approaches could further enhance this model. To validate this hypothesis, we present a 2-stage evolutionary-based path planning algorithm for shepherding a swarm of agents in 2D environments. In the first stage, the algorithm attempts to find the best path for the sheepdog to move from its initial location to a strategic driving location behind the sheep. In the second stage, it calculates and optimises a path for the sheep. It does so by using \emph{way points} on that path as the sequential sub-goals for the sheepdog to aim towards. The proposed algorithm is evaluated in obstacle laden environments via simulation with further improvements achieved.

Nathan K Long, Karl Sammut, Daniel Sgarioto et al.

The simultaneous control of multiple coordinated robotic agents represents an elaborate problem. If solved, however, the interaction between the agents can lead to solutions to sophisticated problems. The concept of swarming, inspired by nature, can be described as the emergence of complex system-level behaviors from the interactions of relatively elementary agents. Due to the effectiveness of solutions found in nature, bio-inspired swarming-based control techniques are receiving a lot of attention in robotics. One method, known as swarm shepherding, is founded on the sheep herding behavior exhibited by sheepdogs, where a swarm of relatively simple agents are governed by a shepherd (or shepherds) which is responsible for high-level guidance and planning. Many studies have been conducted on shepherding as a control technique, ranging from the replication of sheep herding via simulation, to the control of uninhabited vehicles and robots for a variety of applications. We present a comprehensive review of the literature on swarm shepherding to reveal the advantages and potential of the approach to be applied to a plethora of robotic systems in the future.

Nicholas R. Clayton, Hussein Abbass

The design of reward functions in reinforcement learning is a human skill that comes with experience. Unfortunately, there is not any methodology in the literature that could guide a human to design the reward function or to allow a human to transfer the skills developed in designing reward functions to another human and in a systematic manner. In this paper, we use Systematic Instructional Design, an approach in human education, to engineer a machine education methodology to design reward functions for reinforcement learning. We demonstrate the methodology in designing a hierarchical genetic reinforcement learner that adopts a neural network representation to evolve a swarm controller for an agent shepherding a boids-based swarm. The results reveal that the methodology is able to guide the design of hierarchical reinforcement learners, with each model in the hierarchy learning incrementally through a multi-part reward function. The hierarchy acts as a decision fusion function that combines the individual behaviours and skills learnt by each instruction to create a smart shepherd to control the swarm.

Alexander Gee, Hussein Abbass

Swarm control is a difficult problem due to the need to guide a large number of agents simultaneously. We cast the problem as a shepherding problem, similar to biological dogs guiding a group of sheep towards a goal. The shepherd needs to deal with complex and dynamic environments and make decisions in order to direct the swarm from one location to another. In this paper, we design a novel curriculum to teach an artificial intelligence empowered agent to shepherd in the presence of the large state space associated with the shepherding problem and in a transparent manner. The results show that a properly designed curriculum could indeed enhance the speed of learning and the complexity of learnt behaviours.

Hussein Abbass, John Harvey, Kate Yaxley

Artificial Intelligence (AI) technologies could be broadly categorised into Analytics and Autonomy. Analytics focuses on algorithms offering perception, comprehension, and projection of knowledge gleaned from sensorial data. Autonomy revolves around decision making, and influencing and shaping the environment through action production. A smart autonomous system (SAS) combines analytics and autonomy to understand, learn, decide and act autonomously. To be useful, SAS must be trusted and that requires testing. Lifelong learning of a SAS compounds the testing process. In the remote chance that it is possible to fully test and certify the system pre-release, which is theoretically an undecidable problem, it is near impossible to predict the future behaviours that these systems, alone or collectively, will exhibit. While it may be feasible to severely restrict such systems\textquoteright \ learning abilities to limit the potential unpredictability of their behaviours, an undesirable consequence may be severely limiting their utility. In this paper, we propose the architecture for a watchdog AI (WAI) agent dedicated to lifelong functional testing of SAS. We further propose system specifications including a level of abstraction whereby humans shepherd a swarm of WAI agents to oversee an ecosystem made of humans and SAS. The discussion extends to the challenges, pros, and cons of the proposed concept.

Hung Nguyen, Vu Tran, Tung Nguyen et al.

In apprenticeship learning (AL), agents learn by watching or acquiring human demonstrations on some tasks of interest. However, the lack of human demonstrations in novel tasks where they may not be a human expert yet, or when it is too expensive and/or time consuming to acquire human demonstrations motivated a new algorithm: Apprenticeship bootstrapping (ABS). The basic idea is to learn from demonstrations on sub-tasks then autonomously bootstrap a model on the main, more complex, task. The original ABS used inverse reinforcement learning (ABS-IRL). However, the approach is not suitable for continuous action spaces. In this paper, we propose ABS via Deep learning (ABS-DL). It is first validated in a simulation environment on an aerial and ground coordination scenario, where an Unmanned Aerial Vehicle (UAV) is required to maintain three Unmanned Ground Vehicles (UGVs) within a field of view of the UAV 's camera (FoV). Moving a machine learning algorithm from a simulation environment to an actual physical platform is challenging because `mistakes' made by the algorithm while learning could lead to the damage of the platform. We then take this extra step to test the algorithm in a physical environment. We propose a safety-net as a protection layer to ensure that the autonomy of the algorithm in learning does not compromise the safety of the platform. The tests of ABS-DL in the real environment can guarantee a damage-free, collision avoidance behaviour of autonomous bodies. The results show that performance of the proposed approach is comparable to that of a human, and competitive to the traditional approach using expert demonstrations performed on the composite task. The proposed safety-net approach demonstrates its advantages when it enables the UAV to operate more safely under the control of the ABS-DL algorithm.

Aya Hussein, Hussein Abbass

Human-swarm interaction (HSI) involves a number of human factors impacting human behaviour throughout the interaction. As the technologies used within HSI advance, it is more tempting to increase the level of swarm autonomy within the interaction to reduce the workload on humans. Yet, the prospective negative effects of high levels of autonomy on human situational awareness can hinder this process. Flexible autonomy aims at trading-off these effects by changing the level of autonomy within the interaction when required; with mixed-initiatives combining human preferences and automation's recommendations to select an appropriate level of autonomy at a certain point of time. However, the effective implementation of mixed-initiative systems raises fundamental questions on how to combine human preferences and automation recommendations, how to realise the selected level of autonomy, and what the future impacts on the cognitive states of a human are. We explore open challenges that hamper the process of developing effective flexible autonomy. We then highlight the potential benefits of using system modelling techniques in HSI by illustrating how they provide HSI designers with an opportunity to evaluate different strategies for assessing the state of the mission and for adapting the level of autonomy within the interaction to maximise mission success metrics.

Jiangjun Tang, George Leu, Hussein Abbass

Swarm behavior using Boids-like models has been studied primarily using close-proximity spatial sensory information (e.g. vision range). In this study, we propose a novel approach in which the classic definition of boids\textquoteright \ neighborhood that relies on sensory perception and Euclidian space locality is replaced with graph-theoretic network-based proximity mimicking communication and social networks. We demonstrate that networking the boids leads to faster swarming and higher quality of the formation. We further investigate the effect of adversarial learning, whereby an observer attempts to reverse engineer the dynamics of the swarm through observing its behavior. The results show that networking the swarm demonstrated a more robust approach against adversarial learning than a local-proximity neighborhood structure.

George Leu, Hussein Abbass

This paper offers a multi-disciplinary review of knowledge acquisition methods in human activity systems. The review captures the degree of involvement of various types of agencies in the knowledge acquisition process, and proposes a classification with three categories of methods: the human agent, the human-inspired agent, and the autonomous machine agent methods. In the first two categories, the acquisition of knowledge is seen as a cognitive task analysis exercise, while in the third category knowledge acquisition is treated as an autonomous knowledge-discovery endeavour. The motivation for this classification stems from the continuous change over time of the structure, meaning and purpose of human activity systems, which are seen as the factor that fuelled researchers' and practitioners' efforts in knowledge acquisition for more than a century. We show through this review that the KA field is increasingly active due to the higher and higher pace of change in human activity, and conclude by discussing the emergence of a fourth category of knowledge acquisition methods, which are based on red-teaming and co-evolution.

George Leu, Hussein Abbass

In this paper we provide an insight into the skill representation, where skill representation is seen as an essential part of the skill assessment stage in the Computational Red Teaming process. Skill representation is demonstrated in the context of Sudoku puzzle, for which the real human skills used in Sudoku solving, along with their acquisition, are represented computationally in a cognitively plausible manner, by using feed-forward neural networks with back-propagation, and supervised learning. The neural network based skills are then coupled with a hard-coded constraint propagation computational Sudoku solver, in which the solving sequence is kept hard-coded, and the skills are represented through neural networks. The paper demonstrates that the modified solver can achieve different levels of proficiency, depending on the amount of skills acquired through the neural networks. Results are encouraging for developing more complex skill and skill acquisition models usable in general frameworks related to the skill assessment aspect of Computational Red Teaming.

George Leu, Jiangjun Tang, Hussein Abbass

Working memory accounts for the ability of humans to perform cognitive processing, by handling both the representation of information (the mental picture forming the situation awareness) and the space required for processing these information (skill processing). The more complex the skills are, the more processing space they require, the less space becomes available for storage of information. This interplay between situation awareness and skills is critical in many applications. Theoretically, it is less understood in cognition and neuroscience. In the meantime, and practically, it is vital when analysing the mental processes involved in safety-critical domains. In this paper, we use the Sudoku game as a vehicle to study this trade-off. This game combines two features that are present during a user interaction with a software in many safety critical domains: scanning for information and processing of information. We use a society of agents for investigating how this trade-off influences player's proficiency.

Jiangjun Tang, Eleni Petraki, Hussein Abbass

Sociotechnical systems are complex systems, where nonlinear interaction among different players can obscure causal relationships. The absence of mechanisms to help us understand how to create a change in the system makes it hard to manage these systems. Influencing and shaping are social operators acting on sociotechnical systems to design a change. However, the two operators are usually discussed in an ad-hoc manner, without proper guiding models and metrics which assist in adopting these models successfully. Moreover, both social operators rely on accurate understanding of the concept of trust. Without such understanding, neither of these operators can create the required level to create a change in a desirable direction. In this paper, we define these concepts in a concise manner suitable for modelling the concepts and understanding their dynamics. We then introduce a model for influencing and shaping and use Computational Red Teaming principles to design and demonstrate how this model operates. We validate the results computationally through a simulation environment to show social influencing and shaping in an artificial society.