Kate Highnam

In the network security arms race, the defender is significantly disadvantaged as they need to successfully detect and counter every malicious attack. In contrast, the attacker needs to succeed only once. To level the playing field, we investigate the effectiveness of autonomous agents in a realistic network defence scenario. We first outline the problem, provide the background on reinforcement learning and detail our proposed agent design. Using a network environment simulation, with 13 hosts spanning 3 subnets, we train a novel reinforcement learning agent and show that it can reliably defend continual attacks by two advanced persistent threat (APT) red agents: one with complete knowledge of the network layout and another which must discover resources through exploration but is more general.

The Athena Swan Charter is a globally applied framework for fostering and maintaining gender equality within higher education (HE) and research. The original aims of the charter (2005–2015) were to acknowledge commitments to, encourage initiatives for, and recognize achievements in advancing women’s careers in science, technology, engineering, mathematics, and medicine (STEMM) subjects. The charter has evolved and in 2015, Athena Swan was extended to recognize commitment to gender equality across all disciplines, professional and support staff, as well as intersections of gender and race, and trans inclusion. In 2021, following an independent review, the transformed Charter was launched including a commitment to transform gender equality for all gender identities, academic disciplines, and in professional, technical and operational (PTO) directorates. Consequently, criteria for the award of a Charter Mark include recognition of and plans for addressing priority issues, progress toward and success in achieving and transforming gender equality, and, at the highest level of award, contributions to best practice and supporting others.

Communication networks able to withstand hostile environments are critically important for disaster relief operations. In this paper, we consider a challenging scenario where drones have been compromised in the supply chain, during their manufacture, and harbour malicious software capable of wide-ranging and infectious disruption. We investigate multi-agent deep reinforcement learning as a tool for learning defensive strategies that maximise communications bandwidth despite continual adversarial interference. Using a public challenge for learning network resilience strategies, we propose a state-of-the-art symbolic technique and study its superiority over deep reinforcement learning agents. Correspondingly, we identify three specific methods for improving the performance of our neural agents: (1) ensuring each observation contains the necessary information, (2) using symbolic agents to provide a curriculum for learning, and (3) paying close attention to reward. We apply our methods and present a new mixed strategy enabling symbolic and neural agents to work together and improve on all prior results.

Botnets and malware continue to avoid detection by static rules engines when using domain generation algorithms (DGAs) for callouts to unique, dynamically generated web addresses. Common DGA detection techniques fail to reliably detect DGA variants that combine random dictionary words to create domain names that closely mirror legitimate domains. To combat this, we created a novel hybrid neural network, Bilbo the `bagging` model, that analyses domains and scores the likelihood they are generated by such algorithms and therefore are potentially malicious. Bilbo is the first parallel usage of a convolutional neural network (CNN) and a long short-term memory (LSTM) network for DGA detection. Our unique architecture is found to be the most consistent in performance in terms of AUC, F1 score, and accuracy when generalising across different dictionary DGA classification tasks compared to current state-of-the-art deep learning architectures. We validate using reverse-engineered dictionary DGA domains and detail our real-time implementation strategy for scoring real-world network logs within a large financial enterprise. In four hours of actual network traffic, the model discovered at least five potential command-and-control networks that commercial vendor tools did not flag.