Anil A. Bharath

Konstantinos Ntagiantas, Eduardo Pignatelli, Nicholas S. Peters et al.

Atrial Fibrillation (AF) is characterized by disorganised electrical activity in the atria and is known to be sustained by the presence of regions of fibrosis (scars) or functional cellular remodeling, both of which may lead to areas of slow conduction. Estimating the effective conductivity of the myocardium and identifying regions of abnormal propagation is therefore crucial for the effective treatment of AF. We hypothesise that the spatial distribution of tissue conductivity can be directly inferred from an array of concurrently acquired contact electrograms (EGMs). We generate a dataset of simulated cardiac AP propagation using randomised scar distributions and a phenomenological cardiac model and calculate contact EGMs at various positions on the field. EGMs are enriched with noise extracted from biological data acquired in the lab. A deep neural network, based on a modified U-net architecture, is trained to estimate the location of the scar and quantify conductivity of the tissue with a Jaccard index of 91%. We adapt a wavelet-based surrogate testing analysis to confirm that the inferred conductivity distribution is an accurate representation of the ground truth input to the model. We find that the root mean square error (RMSE) between the ground truth and our predictions is significantly smaller ($p_{val}<0.01$) than the RMSE between the ground truth and surrogate samples.

Mario Lino, Stathi Fotiadis, Anil A. Bharath et al.

Numerical simulators are essential tools in the study of natural fluid-systems, but their performance often limits application in practice. Recent machine-learning approaches have demonstrated their ability to accelerate spatio-temporal predictions, although, with only moderate accuracy in comparison. Here we introduce MultiScaleGNN, a novel multi-scale graph neural network model for learning to infer unsteady continuum mechanics in problems encompassing a range of length scales and complex boundary geometries. We demonstrate this method on advection problems and incompressible fluid dynamics, both fundamental phenomena in oceanic and atmospheric processes. Our results show good extrapolation to new domain geometries and parameters for long-term temporal simulations. Simulations obtained with MultiScaleGNN are between two and four orders of magnitude faster than those on which it was trained.

Fatmatulzehra Uslu, Anil A. Bharath

Recently, deep networks have shown impressive performance for the segmentation of cardiac Magnetic Resonance Imaging (MRI) images. However, their achievement is proving slow to transition to widespread use in medical clinics because of robustness issues leading to low trust of clinicians to their results. Predicting run-time quality of segmentation masks can be useful to warn clinicians against poor results. Despite its importance, there are few studies on this problem. To address this gap, we propose a quality control method based on the agreement across decoders of a multi-view network, TMS-Net, measured by the cosine similarity. The network takes three view inputs resliced from the same 3D image along different axes. Different from previous multi-view networks, TMS-Net has a single encoder and three decoders, leading to better noise robustness, segmentation performance and run-time quality estimation in our experiments on the segmentation of the left atrium on STACOM 2013 and STACOM 2018 challenge datasets. We also present a way to generate poor segmentation masks by using noisy images generated with engineered noise and Rician noise to simulate undertraining, high anisotropy and poor imaging settings problems. Our run-time quality estimation method show a good classification of poor and good quality segmentation masks with an AUC reaching to 0.97 on STACOM 2018. We believe that TMS-Net and our run-time quality estimation method has a high potential to increase the thrust of clinicians to automatic image analysis tools.

Christoforos Galazis, Huiyi Wu, Zhuoyu Li et al.

Assessing the structure and function of the right ventricle (RV) is important in the diagnosis of several cardiac pathologies. However, it remains more challenging to segment the RV than the left ventricle (LV). In this paper, we focus on segmenting the RV in both short (SA) and long-axis (LA) cardiac MR images simultaneously. For this task, we propose a new multi-input/output architecture, hybrid 2D/3D geometric spatial TransformEr Multi-Pass fEature pyRAmid (Tempera). Our feature pyramid extends current designs by allowing not only a multi-scale feature output but multi-scale SA and LA input images as well. Tempera transfers learned features between SA and LA images via layer weight sharing and incorporates a geometric target transformer to map the predicted SA segmentation to LA space. Our model achieves an average Dice score of 0.836 and 0.798 for the SA and LA, respectively, and 26.31 mm and 31.19 mm Hausdorff distances. This opens up the potential for the incorporation of RV segmentation models into clinical workflows.

Mario Lino, Stati Fotiadis, Anil A. Bharath et al.

Numerical simulation is an essential tool in many areas of science and engineering, but its performance often limits application in practice or when used to explore large parameter spaces. On the other hand, surrogate deep learning models, while accelerating simulations, often exhibit poor accuracy and ability to generalise. In order to improve these two factors, we introduce REMuS-GNN, a rotation-equivariant multi-scale model for simulating continuum dynamical systems encompassing a range of length scales. REMuS-GNN is designed to predict an output vector field from an input vector field on a physical domain discretised into an unstructured set of nodes. Equivariance to rotations of the domain is a desirable inductive bias that allows the network to learn the underlying physics more efficiently, leading to improved accuracy and generalisation compared with similar architectures that lack such symmetry. We demonstrate and evaluate this method on the incompressible flow around elliptical cylinders.

Christoforos Galazis, Anil Anthony Bharath, Marta Varela

Functional analysis of the left atrium (LA) plays an increasingly important role in the prognosis and diagnosis of cardiovascular diseases. Echocardiography-based measurements of LA dimensions and strains are useful biomarkers, but they provide an incomplete picture of atrial deformations. High-resolution dynamic magnetic resonance images (Cine MRI) offer the opportunity to examine LA motion and deformation in 3D, at higher spatial resolution and with full LA coverage. However, there are no dedicated tools to automatically characterise LA motion in 3D. Thus, we propose a tool that automatically segments the LA and extracts the displacement fields across the cardiac cycle. The pipeline is able to accurately track the LA wall across the cardiac cycle with an average Hausdorff distance of $2.51 \pm 1.3~mm$ and Dice score of $0.96 \pm 0.02$.

Christoforos Galazis, Ching-En Chiu, Tomoki Arichi et al.

Arterial spin labeling (ASL) magnetic resonance imaging (MRI) enables cerebral perfusion measurement, which is crucial in detecting and managing neurological issues in infants born prematurely or after perinatal complications. However, cerebral blood flow (CBF) estimation in infants using ASL remains challenging due to the complex interplay of network physiology, involving dynamic interactions between cardiac output and cerebral perfusion, as well as issues with parameter uncertainty and data noise. We propose a new spatial uncertainty-based physics-informed neural network (PINN), SUPINN, to estimate CBF and other parameters from infant ASL data. SUPINN employs a multi-branch architecture to concurrently estimate regional and global model parameters across multiple voxels. It computes regional spatial uncertainties to weigh the signal. SUPINN can reliably estimate CBF (relative error $-0.3 \pm 71.7$), bolus arrival time (AT) ($30.5 \pm 257.8$), and blood longitudinal relaxation time ($T_{1b}$) ($-4.4 \pm 28.9$), surpassing parameter estimates performed using least squares or standard PINNs. Furthermore, SUPINN produces physiologically plausible spatially smooth CBF and AT maps. Our study demonstrates the successful modification of PINNs for accurate multi-parameter perfusion estimation from noisy and limited ASL data in infants. Frameworks like SUPINN have the potential to advance our understanding of the complex cardio-brain network physiology, aiding in the detection and management of diseases. Source code is provided at: https://github.com/cgalaz01/supinn.

Christoforos Galazis, Samuel Shepperd, Emma Brouwer et al.

The functional analysis of the left atrium (LA) is important for evaluating cardiac health and understanding diseases like atrial fibrillation. Cine MRI is ideally placed for the detailed 3D characterization of LA motion and deformation but is lacking appropriate acquisition and analysis tools. Here, we propose tools for the Analysis for Left Atrial Displacements and DeformatIons using online learning neural Networks (Aladdin) and present a technical feasibility study on how Aladdin can characterize 3D LA function globally and regionally. Aladdin includes an online segmentation and image registration network, and a strain calculation pipeline tailored to the LA. We create maps of LA Displacement Vector Field (DVF) magnitude and LA principal strain values from images of 10 healthy volunteers and 8 patients with cardiovascular disease (CVD), of which 2 had large left ventricular ejection fraction (LVEF) impairment. We additionally create an atlas of these biomarkers using the data from the healthy volunteers. Results showed that Aladdin can accurately track the LA wall across the cardiac cycle and characterize its motion and deformation. Global LA function markers assessed with Aladdin agree well with estimates from 2D Cine MRI. A more marked active contraction phase was observed in the healthy cohort, while the CVD LVEF group showed overall reduced LA function. Aladdin is uniquely able to identify LA regions with abnormal deformation metrics that may indicate focal pathology. We expect Aladdin to have important clinical applications as it can non-invasively characterize atrial pathophysiology. All source code and data are available at: https://github.com/cgalaz01/aladdin_cmr_la.

Antonia Creswell, Anil A Bharath

Generative adversarial networks (GANs) learn a deep generative model that is able to synthesise novel, high-dimensional data samples. New data samples are synthesised by passing latent samples, drawn from a chosen prior distribution, through the generative model. Once trained, the latent space exhibits interesting properties, that may be useful for down stream tasks such as classification or retrieval. Unfortunately, GANs do not offer an "inverse model", a mapping from data space back to latent space, making it difficult to infer a latent representation for a given data sample. In this paper, we introduce a technique, inversion, to project data samples, specifically images, to the latent space using a pre-trained GAN. Using our proposed inversion technique, we are able to identify which attributes of a dataset a trained GAN is able to model and quantify GAN performance, based on a reconstruction loss. We demonstrate how our proposed inversion technique may be used to quantitatively compare performance of various GAN models trained on three image datasets. We provide code for all of our experiments, https://github.com/ToniCreswell/InvertingGAN.

Viktor Schlegel, Anil A Bharath, Zilong Zhao et al.

Privacy-preserving synthetic data offers a promising solution to harness segregated data in high-stakes domains where information is compartmentalized for regulatory, privacy, or institutional reasons. This survey provides a comprehensive framework for understanding the landscape of privacy-preserving synthetic data, presenting the theoretical foundations of generative models and differential privacy followed by a review of state-of-the-art methods across tabular data, images, and text. Our synthesis of evaluation approaches highlights the fundamental trade-off between utility for down-stream tasks and privacy guarantees, while identifying critical research gaps: the lack of realistic benchmarks representing specialized domains and insufficient empirical evaluations required to contextualise formal guarantees. Through empirical analysis of four leading methods on five real-world datasets from specialized domains, we demonstrate significant performance degradation under realistic privacy constraints ($ε\leq 4$), revealing a substantial gap between results reported on general domain benchmarks and performance on domain-specific data. %Our findings highlight key challenges including unaccounted privacy leakage, insufficient empirical verification of formal guarantees, and a critical deficit of realistic benchmarks. These challenges underscore the need for robust evaluation frameworks, standardized benchmarks for specialized domains, and improved techniques to address the unique requirements of privacy-sensitive fields such that this technology can deliver on its considerable potential.

Paul Pöhl, Viktor Schlegel, Hao Li et al.

Generating synthetic ECG data has numerous applications in healthcare, from educational purposes to simulating scenarios and forecasting trends. While recent diffusion models excel at generating short ECG segments, they struggle with longer sequences needed for many clinical applications. This paper proposes a novel three-layer synthesis framework for generating realistic long-form ECG signals. We first generate high-fidelity single beats using a diffusion model, then synthesize inter-beat features preserving critical temporal dependencies, and finally assemble beats into coherent long sequences using feature-guided matching. Our comprehensive evaluation demonstrates that the resulting synthetic ECGs maintain both beat-level morphological fidelity and clinically relevant inter-beat relationships. In arrhythmia classification tasks, our long-form synthetic ECGs significantly outperform end-to-end long-form ECG generation using the diffusion model, highlighting their potential for increasing utility for downstream applications. The approach enables generation of unprecedented multi-minute ECG sequences while preserving essential diagnostic characteristics.

Yuping Wu, Viktor Schlegel, Warren Del-Pinto et al.

Training data is fundamental to the success of modern machine learning models, yet in high-stakes domains such as healthcare, the use of real-world training data is severely constrained by concerns over privacy leakage. A promising solution to this challenge is the use of differentially private (DP) synthetic data, which offers formal privacy guarantees while maintaining data utility. However, striking the right balance between privacy protection and utility remains challenging in clinical note synthesis, given its domain specificity and the complexity of long-form text generation. In this paper, we present Term2Note, a methodology to synthesise long clinical notes under strong DP constraints. By structurally separating content and form, Term2Note generates section-wise note content conditioned on DP medical terms, with each governed by separate DP constraints. A DP quality maximiser further enhances synthetic notes by selecting high-quality outputs. Experimental results show that Term2Note produces synthetic notes with statistical properties closely aligned with real clinical notes, demonstrating strong fidelity. In addition, multi-label classification models trained on these synthetic notes perform comparably to those trained on real data, confirming their high utility. Compared to existing DP text generation baselines, Term2Note achieves substantial improvements in both fidelity and utility while operating under fewer assumptions, suggesting its potential as a viable privacy-preserving alternative to using sensitive clinical notes.

Yidan Sun, Viktor Schlegel, Srinivasan Nandakumar et al.

Generative AI offers transformative potential for high-stakes domains such as healthcare and finance, yet privacy and regulatory barriers hinder the use of real-world data. To address this, differentially private synthetic data generation has emerged as a promising alternative. In this work, we introduce a unified benchmark to systematically evaluate the utility and fidelity of text datasets generated under formal Differential Privacy (DP) guarantees. Our benchmark addresses key challenges in domain-specific benchmarking, including choice of representative data and realistic privacy budgets, accounting for pre-training and a variety of evaluation metrics. We assess state-of-the-art privacy-preserving generation methods across five domain-specific datasets, revealing significant utility and fidelity degradation compared to real data, especially under strict privacy constraints. These findings underscore the limitations of current approaches, outline the need for advanced privacy-preserving data sharing methods and set a precedent regarding their evaluation in realistic scenarios.

Yidan Sun, Viktor Schlegel, Srinivasan Nandakumar et al.

Data-driven decision support in high-stakes domains like healthcare and finance faces significant barriers to data sharing due to regulatory, institutional, and privacy concerns. While recent generative AI models, such as large language models, have shown impressive performance in open-domain tasks, their adoption in sensitive environments remains limited by unpredictable behaviors and insufficient privacy-preserving datasets for benchmarking. Existing anonymization methods are often inadequate, especially for unstructured text, as redaction and masking can still allow re-identification. Differential Privacy (DP) offers a principled alternative, enabling the generation of synthetic data with formal privacy assurances. In this work, we address these challenges through three key contributions. First, we introduce a comprehensive evaluation framework with standardized utility and fidelity metrics, encompassing nine curated datasets that capture domain-specific complexities such as technical jargon, long-context dependencies, and specialized document structures. Second, we conduct a large-scale empirical study benchmarking state-of-the-art DP text generation methods and LLMs of varying sizes and different fine-tuning strategies, revealing that high-quality domain-specific synthetic data generation under DP constraints remains an unsolved challenge, with performance degrading as domain complexity increases. Third, we develop a membership inference attack (MIA) methodology tailored for synthetic text, providing first empirical evidence that the use of public datasets - potentially present in pre-training corpora - can invalidate claimed privacy guarantees. Our findings underscore the urgent need for rigorous privacy auditing and highlight persistent gaps between open-domain and specialist evaluations, informing responsible deployment of generative AI in privacy-sensitive, high-stakes settings.

Vivek Mohan, Biyan Zhou, Zhou Wang et al.

This work presents an efficient decoding pipeline for neuromorphic implantable brain-machine interfaces (Neu-iBMI), leveraging sparse neural event data from an event-based neural sensing scheme. We introduce a tunable event filter (EvFilter), which also functions as a spike detector (EvFilter-SPD), significantly reducing the number of events processed for decoding by 192X and 554X, respectively. The proposed pipeline achieves high decoding performance, up to R^2=0.73, with ANN- and SNN-based decoders, eliminating the need for signal recovery, spike detection, or sorting, commonly performed in conventional iBMI systems. The SNN-Decoder reduces computations and memory required by 5-23X compared to NN-, and LSTM-Decoders, while the ST-NN-Decoder delivers similar performance to an LSTM-Decoder requiring 2.5X fewer resources. This streamlined approach significantly reduces computational and memory demands, making it ideal for low-power, on-implant, or wearable iBMIs.

Stathi Fotiadis, Mario Lino, Shunlong Hu et al.

Deep neural networks have become increasingly of interest in dynamical system prediction, but out-of-distribution generalization and long-term stability still remains challenging. In this work, we treat the domain parameters of dynamical systems as factors of variation of the data generating process. By leveraging ideas from supervised disentanglement and causal factorization, we aim to separate the domain parameters from the dynamics in the latent space of generative models. In our experiments we model dynamics both in phase space and in video sequences and conduct rigorous OOD evaluations. Results indicate that disentangled VAEs adapt better to domain parameters spaces that were not present in the training data. At the same time, disentanglement can improve the long-term and out-of-distribution predictions of state-of-the-art models in video sequences.

Tianhong Dai, Hengyan Liu, Kai Arulkumaran et al.

Hindsight experience replay (HER) is a goal relabelling technique typically used with off-policy deep reinforcement learning algorithms to solve goal-oriented tasks; it is well suited to robotic manipulation tasks that deliver only sparse rewards. In HER, both trajectories and transitions are sampled uniformly for training. However, not all of the agent's experiences contribute equally to training, and so naive uniform sampling may lead to inefficient learning. In this paper, we propose diversity-based trajectory and goal selection with HER (DTGSH). Firstly, trajectories are sampled according to the diversity of the goal states as modelled by determinantal point processes (DPPs). Secondly, transitions with diverse goal states are selected from the trajectories by using k-DPPs. We evaluate DTGSH on five challenging robotic manipulation tasks in simulated robot environments, where we show that our method can learn more quickly and reach higher performance than other state-of-the-art approaches on all tasks.

Mario Lino, Chris Cantwell, Anil A. Bharath et al.

Continuum mechanics simulators, numerically solving one or more partial differential equations, are essential tools in many areas of science and engineering, but their performance often limits application in practice. Recent modern machine learning approaches have demonstrated their ability to accelerate spatio-temporal predictions, although, with only moderate accuracy in comparison. Here we introduce MultiScaleGNN, a novel multi-scale graph neural network model for learning to infer unsteady continuum mechanics. MultiScaleGNN represents the physical domain as an unstructured set of nodes, and it constructs one or more graphs, each of them encoding different scales of spatial resolution. Successive learnt message passing between these graphs improves the ability of GNNs to capture and forecast the system state in problems encompassing a range of length scales. Using graph representations, MultiScaleGNN can impose periodic boundary conditions as an inductive bias on the edges in the graphs, and achieve independence to the nodes' positions. We demonstrate this method on advection problems and incompressible fluid dynamics. Our results show that the proposed model can generalise from uniform advection fields to high-gradient fields on complex domains at test time and infer long-term Navier-Stokes solutions within a range of Reynolds numbers. Simulations obtained with MultiScaleGNN are between two and four orders of magnitude faster than the ones on which it was trained.

Mario Lino, Chris Cantwell, Stathi Fotiadis et al.

We investigate the performance of fully convolutional networks to simulate the motion and interaction of surface waves in open and closed complex geometries. We focus on a U-Net architecture and analyse how well it generalises to geometric configurations not seen during training. We demonstrate that a modified U-Net architecture is capable of accurately predicting the height distribution of waves on a liquid surface within curved and multi-faceted open and closed geometries, when only simple box and right-angled corner geometries were seen during training. We also consider a separate and independent 3D CNN for performing time-interpolation on the predictions produced by our U-Net. This allows generating simulations with a smaller time-step size than the one the U-Net has been trained for.

Tianhong Dai, Hengyan Liu, Anil Anthony Bharath

Episodic self-imitation learning, a novel self-imitation algorithm with a trajectory selection module and an adaptive loss function, is proposed to speed up reinforcement learning. Compared to the original self-imitation learning algorithm, which samples good state-action pairs from the experience replay buffer, our agent leverages entire episodes with hindsight to aid self-imitation learning. A selection module is introduced to filter uninformative samples from each episode of the update. The proposed method overcomes the limitations of the standard self-imitation learning algorithm, a transitions-based method which performs poorly in handling continuous control environments with sparse rewards. From the experiments, episodic self-imitation learning is shown to perform better than baseline on-policy algorithms, achieving comparable performance to state-of-the-art off-policy algorithms in several simulated robot control tasks. The trajectory selection module is shown to prevent the agent learning undesirable hindsight experiences. With the capability of solving sparse reward problems in continuous control settings, episodic self-imitation learning has the potential to be applied to real-world problems that have continuous action spaces, such as robot guidance and manipulation.

Stathi Fotiadis, Eduardo Pignatelli, Mario Lino Valencia et al.

Dynamical systems can be modelled by partial differential equations and numerical computations are used everywhere in science and engineering. In this work, we investigate the performance of recurrent and convolutional deep neural network architectures to predict the surface waves. The system is governed by the Saint-Venant equations. We improve on the long-term prediction over previous methods while keeping the inference time at a fraction of numerical simulations. We also show that convolutional networks perform at least as well as recurrent networks in this task. Finally, we assess the generalisation capability of each network by extrapolating in longer time-frames and in different physical settings.

Tianhong Dai, Kai Arulkumaran, Tamara Gerbert et al.

Deep reinforcement learning has the potential to train robots to perform complex tasks in the real world without requiring accurate models of the robot or its environment. A practical approach is to train agents in simulation, and then transfer them to the real world. One popular method for achieving transferability is to use domain randomisation, which involves randomly perturbing various aspects of a simulated environment in order to make trained agents robust to the reality gap. However, less work has gone into understanding such agents - which are deployed in the real world - beyond task performance. In this work we examine such agents, through qualitative and quantitative comparisons between agents trained with and without visual domain randomisation. We train agents for Fetch and Jaco robots on a visuomotor control task and evaluate how well they generalise using different testing conditions. Finally, we investigate the internals of the trained agents by using a suite of interpretability techniques. Our results show that the primary outcome of domain randomisation is more robust, entangled representations, accompanied with larger weights with greater spatial structure; moreover, the types of changes are heavily influenced by the task setup and presence of additional proprioceptive inputs. Additionally, we demonstrate that our domain randomised agents require higher sample complexity, can overfit and more heavily rely on recurrent processing. Furthermore, even with an improved saliency method introduced in this work, we show that qualitative studies may not always correspond with quantitative measures, necessitating the combination of inspection tools in order to provide sufficient insights into the behaviour of trained agents.

Marta Sarrico, Kai Arulkumaran, Andrea Agostinelli et al.

Deep networks have enabled reinforcement learning to scale to more complex and challenging domains, but these methods typically require large quantities of training data. An alternative is to use sample-efficient episodic control methods: neuro-inspired algorithms which use non-/semi-parametric models that predict values based on storing and retrieving previously experienced transitions. One way to further improve the sample efficiency of these approaches is to use more principled exploration strategies. In this work, we therefore propose maximum entropy mellowmax episodic control (MEMEC), which samples actions according to a Boltzmann policy with a state-dependent temperature. We demonstrate that MEMEC outperforms other uncertainty- and softmax-based exploration methods on classic reinforcement learning environments and Atari games, achieving both more rapid learning and higher final rewards.

Andrea Agostinelli, Kai Arulkumaran, Marta Sarrico et al.

Recently, neuro-inspired episodic control (EC) methods have been developed to overcome the data-inefficiency of standard deep reinforcement learning approaches. Using non-/semi-parametric models to estimate the value function, they learn rapidly, retrieving cached values from similar past states. In realistic scenarios, with limited resources and noisy data, maintaining meaningful representations in memory is essential to speed up the learning and avoid catastrophic forgetting. Unfortunately, EC methods have a large space and time complexity. We investigate different solutions to these problems based on prioritising and ranking stored states, as well as online clustering techniques. We also propose a new dynamic online k-means algorithm that is both computationally-efficient and yields significantly better performance at smaller memory sizes; we validate this approach on classic reinforcement learning environments and Atari games.

Wilhelm E. Sorteberg, Stef Garasto, Alison S. Pouplin et al.

Humans gain an implicit understanding of physical laws through observing and interacting with the world. Endowing an autonomous agent with an understanding of physical laws through experience and observation is seldom practical: we should seek alternatives. Fortunately, many of the laws of behaviour of the physical world can be derived from prior knowledge of dynamical systems, expressed through the use of partial differential equations. In this work, we suggest a neural network capable of understanding a specific physical phenomenon: wave propagation in a two-dimensional medium. We define `understanding' in this context as the ability to predict the future evolution of the spatial patterns of rendered wave amplitude from a relatively small set of initial observations. The inherent complexity of the wave equations -- together with the existence of reflections and interference -- makes the prediction problem non-trivial. A network capable of making approximate predictions also unlocks the opportunity to speed-up numerical simulations for wave propagation. To this aim, we created a novel dataset of simulated wave motion and built a predictive deep neural network comprising of three main blocks: an encoder, a propagator made by 3 LSTMs, and a decoder. Results show reasonable predictions for as long as 80 time steps into the future on a dataset not seen during training. Furthermore, the network is able to generalize to an initial condition that is qualitatively different from those seen during training.

Chris D. Cantwell, Yumnah Mohamied, Konstantinos N. Tzortzis et al.

We review some of the latest approaches to analysing cardiac electrophysiology data using machine learning and predictive modelling. Cardiac arrhythmias, particularly atrial fibrillation, are a major global healthcare challenge. Treatment is often through catheter ablation, which involves the targeted localized destruction of regions of the myocardium responsible for initiating or perpetuating the arrhythmia. Ablation targets are either anatomically defined, or identified based on their functional properties as determined through the analysis of contact intracardiac electrograms acquired with increasing spatial density by modern electroanatomic mapping systems. While numerous quantitative approaches have been investigated over the past decades for identifying these critical curative sites, few have provided a reliable and reproducible advance in success rates. Machine learning techniques, including recent deep-learning approaches, offer a potential route to gaining new insight from this wealth of highly complex spatio-temporal information that existing methods struggle to analyse. Coupled with predictive modelling, these techniques offer exciting opportunities to advance the field and produce more accurate diagnoses and robust personalised treatment. We outline some of these methods and illustrate their use in making predictions from the contact electrogram and augmenting predictive modelling tools, both by more rapidly predicting future states of the system and by inferring the parameters of these models from experimental observations.

Antonia Creswell, Alison Pouplin, Anil A Bharath

We propose a novel deep learning model for classifying medical images in the setting where there is a large amount of unlabelled medical data available, but labelled data is in limited supply. We consider the specific case of classifying skin lesions as either malignant or benign. In this setting, the proposed approach -- the semi-supervised, denoising adversarial autoencoder -- is able to utilise vast amounts of unlabelled data to learn a representation for skin lesions, and small amounts of labelled data to assign class labels based on the learned representation. We analyse the contributions of both the adversarial and denoising components of the model and find that the combination yields superior classification performance in the setting of limited labelled training data.

Fatmatulzehra Uslu, Anil Anthony Bharath

Demographic studies suggest that changes in the retinal vasculature geometry, especially in vessel width, are associated with the incidence or progression of eye-related or systemic diseases. To date, the main information source for width estimation from fundus images has been the intensity profile between vessel edges. However, there are many factors affecting the intensity profile: pathologies, the central light reflex and local illumination levels, to name a few. In this study, we introduce three information sources for width estimation. These are the probability profiles of vessel interior, centreline and edge locations generated by a deep network. The probability profiles provide direct access to vessel geometry and are used in the likelihood calculation for a Bayesian method, particle filtering. We also introduce a geometric model which can handle non-ideal conditions of the probability profiles. Our experiments conducted on the REVIEW dataset yielded consistent estimates of vessel width, even in cases when one of the vessel edges is difficult to identify. Moreover, our results suggest that the method is better than human observers at locating edges of low contrast vessels.

Antonia Creswell, Yumnah Mohamied, Biswa Sengupta et al.

We propose a novel generative model architecture designed to learn representations for images that factor out a single attribute from the rest of the representation. A single object may have many attributes which when altered do not change the identity of the object itself. Consider the human face; the identity of a particular person is independent of whether or not they happen to be wearing glasses. The attribute of wearing glasses can be changed without changing the identity of the person. However, the ability to manipulate and alter image attributes without altering the object identity is not a trivial task. Here, we are interested in learning a representation of the image that separates the identity of an object (such as a human face) from an attribute (such as 'wearing glasses'). We demonstrate the success of our factorization approach by using the learned representation to synthesize the same face with and without a chosen attribute. We refer to this specific synthesis process as image attribute manipulation. We further demonstrate that our model achieves competitive scores, with state of the art, on a facial attribute classification task.

Antonia Creswell, Anil A. Bharath, Biswa Sengupta

Robustness and security of machine learning (ML) systems are intertwined, wherein a non-robust ML system (classifiers, regressors, etc.) can be subject to attacks using a wide variety of exploits. With the advent of scalable deep learning methodologies, a lot of emphasis has been put on the robustness of supervised, unsupervised and reinforcement learning algorithms. Here, we study the robustness of the latent space of a deep variational autoencoder (dVAE), an unsupervised generative framework, to show that it is indeed possible to perturb the latent space, flip the class predictions and keep the classification probability approximately equal before and after an attack. This means that an agent that looks at the outputs of a decoder would remain oblivious to an attack.

Antonia Creswell, Tom White, Vincent Dumoulin et al.

Generative adversarial networks (GANs) provide a way to learn deep representations without extensively annotated training data. They achieve this through deriving backpropagation signals through a competitive process involving a pair of networks. The representations that can be learned by GANs may be used in a variety of applications, including image synthesis, semantic image editing, style transfer, image super-resolution and classification. The aim of this review paper is to provide an overview of GANs for the signal processing community, drawing on familiar analogies and concepts where possible. In addition to identifying different methods for training and constructing GANs, we also point to remaining challenges in their theory and application.

Antonia Creswell, Kai Arulkumaran, Anil A. Bharath

Denoising autoencoders (DAEs) are powerful deep learning models used for feature extraction, data generation and network pre-training. DAEs consist of an encoder and decoder which may be trained simultaneously to minimise a loss (function) between an input and the reconstruction of a corrupted version of the input. There are two common loss functions used for training autoencoders, these include the mean-squared error (MSE) and the binary cross-entropy (BCE). When training autoencoders on image data a natural choice of loss function is BCE, since pixel values may be normalised to take values in [0,1] and the decoder model may be designed to generate samples that take values in (0,1). We show theoretically that DAEs trained to minimise BCE may be used to take gradient steps in the data space towards regions of high probability under the data-generating distribution. Previously this had only been shown for DAEs trained using MSE. As a consequence of the theory, iterative application of a trained DAE moves a data sample from regions of low probability to regions of higher probability under the data-generating distribution. Firstly, we validate the theory by showing that novel data samples, consistent with the training data, may be synthesised when the initial data samples are random noise. Secondly, we motivate the theory by showing that initial data samples synthesised via other methods may be improved via iterative application of a trained DAE to those initial samples.

Kai Arulkumaran, Marc Peter Deisenroth, Miles Brundage et al.

Deep reinforcement learning is poised to revolutionise the field of AI and represents a step towards building autonomous systems with a higher level understanding of the visual world. Currently, deep learning is enabling reinforcement learning to scale to problems that were previously intractable, such as learning to play video games directly from pixels. Deep reinforcement learning algorithms are also applied to robotics, allowing control policies for robots to be learned directly from camera inputs in the real world. In this survey, we begin with an introduction to the general field of reinforcement learning, then progress to the main streams of value-based and policy-based methods. Our survey will cover central algorithms in deep reinforcement learning, including the deep $Q$-network, trust region policy optimisation, and asynchronous advantage actor-critic. In parallel, we highlight the unique advantages of deep neural networks, focusing on visual understanding via reinforcement learning. To conclude, we describe several current areas of research within the field.

Antonia Creswell, Anil Anthony Bharath

Unsupervised learning is of growing interest because it unlocks the potential held in vast amounts of unlabelled data to learn useful representations for inference. Autoencoders, a form of generative model, may be trained by learning to reconstruct unlabelled input data from a latent representation space. More robust representations may be produced by an autoencoder if it learns to recover clean input samples from corrupted ones. Representations may be further improved by introducing regularisation during training to shape the distribution of the encoded data in latent space. We suggest denoising adversarial autoencoders, which combine denoising and regularisation, shaping the distribution of latent space using adversarial training. We introduce a novel analysis that shows how denoising may be incorporated into the training and sampling of adversarial autoencoders. Experiments are performed to assess the contributions that denoising makes to the learning of representations for classification and sample synthesis. Our results suggest that autoencoders trained using a denoising criterion achieve higher classification performance, and can synthesise samples that are more consistent with the input data than those trained without a corruption process.

Antonia Creswell, Anil Anthony Bharath

Generative adversarial networks (GANs) learn to synthesise new samples from a high-dimensional distribution by passing samples drawn from a latent space through a generative network. When the high-dimensional distribution describes images of a particular data set, the network should learn to generate visually similar image samples for latent variables that are close to each other in the latent space. For tasks such as image retrieval and image classification, it may be useful to exploit the arrangement of the latent space by projecting images into it, and using this as a representation for discriminative tasks. GANs often consist of multiple layers of non-linear computations, making them very difficult to invert. This paper introduces techniques for projecting image samples into the latent space using any pre-trained GAN, provided that the computational graph is available. We evaluate these techniques on both MNIST digits and Omniglot handwritten characters. In the case of MNIST digits, we show that projections into the latent space maintain information about the style and the identity of the digit. In the case of Omniglot characters, we show that even characters from alphabets that have not been seen during training may be projected well into the latent space; this suggests that this approach may have applications in one-shot learning.

Antonia Creswell, Kai Arulkumaran, Anil Anthony Bharath

We focus on generative autoencoders, such as variational or adversarial autoencoders, which jointly learn a generative model alongside an inference model. Generative autoencoders are those which are trained to softly enforce a prior on the latent distribution learned by the inference model. We call the distribution to which the inference model maps observed samples, the learned latent distribution, which may not be consistent with the prior. We formulate a Markov chain Monte Carlo (MCMC) sampling process, equivalent to iteratively decoding and encoding, which allows us to sample from the learned latent distribution. Since, the generative model learns to map from the learned latent distribution, rather than the prior, we may use MCMC to improve the quality of samples drawn from the generative model, especially when the learned latent distribution is far from the prior. Using MCMC sampling, we are able to reveal previously unseen differences between generative autoencoders trained either with or without a denoising criterion.

Christoforos C. Charalambous, Anil A. Bharath

There are several confounding factors that can reduce the accuracy of gait recognition systems. These factors can reduce the distinctiveness, or alter the features used to characterise gait, they include variations in clothing, lighting, pose and environment, such as the walking surface. Full invariance to all confounding factors is challenging in the absence of high-quality labelled training data. We introduce a simulation-based methodology and a subject-specific dataset which can be used for generating synthetic video frames and sequences for data augmentation. With this methodology, we generated a multi-modal dataset. In addition, we supply simulation files that provide the ability to simultaneously sample from several confounding variables. The basis of the data is real motion capture data of subjects walking and running on a treadmill at different speeds. Results from gait recognition experiments suggest that information about the identity of subjects is retained within synthetically generated examples. The dataset and methodology allow studies into fully-invariant identity recognition spanning a far greater number of observation conditions than would otherwise be possible.

Antonia Creswell, Anil A. Bharath

The cost function used to train a generative model should fit the purpose of the model. If the model is intended for tasks such as generating perceptually correct samples, it is beneficial to maximise the likelihood of a sample drawn from the model, Q, coming from the same distribution as the training data, P. This is equivalent to minimising the Kullback-Leibler (KL) distance, KL[Q||P]. However, if the model is intended for tasks such as retrieval or classification it is beneficial to maximise the likelihood that a sample drawn from the training data is captured by the model, equivalent to minimising KL[P||Q]. The cost function used in adversarial training optimises the Jensen-Shannon entropy which can be seen as an even interpolation between KL[Q||P] and KL[P||Q]. Here, we propose an alternative adversarial cost function which allows easy tuning of the model for either task. Our task specific cost function is evaluated on a dataset of hand-written characters in the following tasks: Generation, retrieval and one-shot learning.

Kai Arulkumaran, Nat Dilokthanakul, Murray Shanahan et al.

In this paper we combine one method for hierarchical reinforcement learning - the options framework - with deep Q-networks (DQNs) through the use of different "option heads" on the policy network, and a supervisory network for choosing between the different options. We utilise our setup to investigate the effects of architectural constraints in subtasks with positive and negative transfer, across a range of network capacities. We empirically show that our augmented DQN has lower sample complexity when simultaneously learning subtasks with negative transfer, without degrading performance when learning subtasks with positive transfer.

Jose Rivera-Rubio, Ioannis Alexiou, Anil A. Bharath

Vision is one of the most important of the senses, and humans use it extensively during navigation. We evaluated different types of image and video frame descriptors that could be used to determine distinctive visual landmarks for localizing a person based on what is seen by a camera that they carry. To do this, we created a database containing over 3 km of video-sequences with ground-truth in the form of distance travelled along different corridors. Using this database, the accuracy of localization - both in terms of knowing which route a user is on - and in terms of position along a certain route, can be evaluated. For each type of descriptor, we also tested different techniques to encode visual structure and to search between journeys to estimate a user's position. The techniques include single-frame descriptors, those using sequences of frames, and both colour and achromatic descriptors. We found that single-frame indexing worked better within this particular dataset. This might be because the motion of the person holding the camera makes the video too dependent on individual steps and motions of one particular journey. Our results suggest that appearance-based information could be an additional source of navigational data indoors, augmenting that provided by, say, radio signal strength indicators (RSSIs). Such visual information could be collected by crowdsourcing low-resolution video feeds, allowing journeys made by different users to be associated with each other, and location to be inferred without requiring explicit mapping. This offers a complementary approach to methods based on simultaneous localization and mapping (SLAM) algorithms.