Author name cluster

Roderick Murray-Smith

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

15 papers

2 author rows

NeurIPS Conference 2024 Conference Paper

Generative Fractional Diffusion Models

Gabriel Nobis
Maximilian Springenberg
Marco Aversa
Michael Detzel
Rembert Daems
Roderick Murray-Smith
Shinichi Nakajima
Sebastian Lapuschkin

We introduce the first continuous-time score-based generative model that leverages fractional diffusion processes for its underlying dynamics. Although diffusion models have excelled at capturing data distributions, they still suffer from various limitations such as slow convergence, mode-collapse on imbalanced data, and lack of diversity. These issues are partially linked to the use of light-tailed Brownian motion (BM) with independent increments. In this paper, we replace BM with an approximation of its non-Markovian counterpart, fractional Brownian motion (fBM), characterized by correlated increments and Hurst index $H \in (0, 1)$, where $H=0. 5$ recovers the classical BM. To ensure tractable inference and learning, we employ a recently popularized Markov approximation of fBM (MA-fBM) and derive its reverse-time model, resulting in *generative fractional diffusion models* (GFDM). We characterize the forward dynamics using a continuous reparameterization trick and propose *augmented score matching* to efficiently learn the score function, which is partly known in closed form, at minimal added cost. The ability to drive our diffusion model via MA-fBM offers flexibility and control. $H \leq 0. 5$ enters the regime of *rough paths* whereas $H>0. 5$ regularizes diffusion paths and invokes long-term memory. The Markov approximation allows added control by varying the number of Markov processes linearly combined to approximate fBM. Our evaluations on real image datasets demonstrate that GFDM achieves greater pixel-wise diversity and enhanced image quality, as indicated by a lower FID, offering a promising alternative to traditional diffusion models

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

Is One GPU Enough? Pushing Image Generation at Higher-Resolutions with Foundation Models.

Athanasios Tragakis
Marco Aversa
Chaitanya Kaul
Roderick Murray-Smith
Daniele Faccio

In this work, we introduce Pixelsmith, a zero-shot text-to-image generative framework to sample images at higher resolutions with a single GPU. We are the first to show that it is possible to scale the output of a pre-trained diffusion model by a factor of 1000, opening the road to gigapixel image generation at no extra cost. Our cascading method uses the image generated at the lowest resolution as baseline to sample at higher resolutions. For the guidance, we introduce the Slider, a mechanism that fuses the overall structure contained in the first-generated image with enhanced fine details. At each inference step, we denoise patches rather than the entire latent space, minimizing memory demands so that a single GPU can handle the process, regardless of the image's resolution. Our experimental results show that this method not only achieves higher quality and diversity compared to existing techniques but also reduces sampling time and ablation artifacts.

PDF Details DOI

TMLR Journal 2023 Journal Article

Data Models for Dataset Drift Controls in Machine Learning With Optical Images

Luis Oala
Marco Aversa
Gabriel Nobis
Kurt Willis
Yoan Neuenschwander
Michèle Buck
Christian Matek
Jerome Extermann

Camera images are ubiquitous in machine learning research. They also play a central role in the delivery of important public services spanning medicine or environmental surveying. However, the application of machine learning models in these domains has been limited because of robustness concerns. A primary failure mode are performance drops due to differences between the training and deployment data. While there are methods to prospectively validate the robustness of machine learning models to such dataset drifts, existing approaches do not account for explicit models of machine learning's primary object of interest: the data. This limits our ability to study and understand the relationship between data generation and downstream machine learning model performance in a physically accurate manner. In this study, we demonstrate how to overcome this limitation by pairing traditional machine learning with physical optics to obtain explicit and differentiable data models. We demonstrate how such data models can be constructed for image data and used to control downstream machine learning model performance related to dataset drift. The findings are distilled into three applications. First, drift synthesis enables the controlled generation of physically faithful drift test cases to power model selection and targeted generalization. Second, the gradient connection between machine learning task model and data model allows advanced, precise tolerancing of task model sensitivity to changes in the data generation. These drift forensics can be used to precisely specify the acceptable data environments in which a task model may be run. Third, drift optimization opens up the possibility to create drifts that can help the task model learn better faster, effectively optimizing the data generating process itself to support the downstream machine vision task. This is an interesting upgrade to existing imaging pipelines which traditionally have been optimized to be consumed by human users but not machine learning models. Alongside the data model code we release two datasets to the public that we collected as part of this work. In total, the two datasets, Raw-Microscopy and Raw-Drone, comprise 1,488 scientifically calibrated reference raw sensor measurements, 8,928 raw intensity variations as well as 17,856 images processed through twelve data models with different configurations. A guide to access the open code and datasets is available at https://github.com/aiaudit-org/raw2logit.

PDF Details

NeurIPS Conference 2023 Conference Paper

DiffInfinite: Large Mask-Image Synthesis via Parallel Random Patch Diffusion in Histopathology

Marco Aversa
Gabriel Nobis
Miriam Hägele
Kai Standvoss
Mihaela Chirica
Roderick Murray-Smith
Ahmed M. Alaa
Lukas Ruff

We present DiffInfinite, a hierarchical diffusion model that generates arbitrarily large histological images while preserving long-range correlation structural information. Our approach first generates synthetic segmentation masks, subsequently used as conditions for the high-fidelity generative diffusion process. The proposed sampling method can be scaled up to any desired image size while only requiring small patches for fast training. Moreover, it can be parallelized more efficiently than previous large-content generation methods while avoiding tiling artifacts. The training leverages classifier-free guidance to augment a small, sparsely annotated dataset with unlabelled data. Our method alleviates unique challenges in histopathological imaging practice: large-scale information, costly manual annotation, and protective data handling. The biological plausibility of DiffInfinite data is evaluated in a survey by ten experienced pathologists as well as a downstream classification and segmentation task. Samples from the model score strongly on anti-copying metrics which is relevant for the protection of patient data.

PDF Details

TMLR Journal 2023 Journal Article

Subgraph Permutation Equivariant Networks

Joshua Mitton
Roderick Murray-Smith

In this work we develop a new method, named Sub-graph Permutation Equivariant Networks (SPEN), which provides a framework for building graph neural networks that operate on sub-graphs, while using a base update function that is permutation equivariant, that are equivariant to a novel choice of automorphism group. Message passing neural networks have been shown to be limited in their expressive power and recent approaches to over come this either lack scalability or require structural information to be encoded into the feature space. The general framework presented here overcomes the scalability issues associated with global permutation equivariance by operating more locally on sub-graphs. In addition, through operating on sub-graphs the expressive power of higher-dimensional global permutation equivariant networks is improved; this is due to fact that two non-distinguishable graphs often contain distinguishable sub-graphs. Furthermore, the proposed framework only requires a choice of $k$-hops for creating ego-network sub-graphs and a choice of representation space to be used for each layer, which makes the method easily applicable across a range of graph based domains. We experimentally validate the method on a range of graph benchmark classification tasks, demonstrating statistically indistinguishable results from the state-of-the-art on six out of seven benchmarks. Further, we demonstrate that the use of local update functions offers a significant improvement in GPU memory over global methods.

PDF Details

NeurIPS Conference 2022 Conference Paper

Bessel Equivariant Networks for Inversion of Transmission Effects in Multi-Mode Optical Fibres

Joshua Mitton
Simon Mekhail
Miles Padgett
Daniele Faccio
Marco Aversa
Roderick Murray-Smith

We develop a new type of model for solving the task of inverting the transmission effects of multi-mode optical fibres through the construction of an $\mathrm{SO}^{+}(2, 1)$-equivariant neural network. This model takes advantage of the of the azimuthal correlations known to exist in fibre speckle patterns and naturally accounts for the difference in spatial arrangement between input and speckle patterns. In addition, we use a second post-processing network to remove circular artifacts, fill gaps, and sharpen the images, which is required due to the nature of optical fibre transmission. This two stage approach allows for the inspection of the predicted images produced by the more robust physically motivated equivariant model, which could be useful in a safety-critical application, or by the output of both models, which produces high quality images. Further, this model can scale to previously unachievable resolutions of imaging with multi-mode optical fibres and is demonstrated on $256 \times 256$ pixel images. This is a result of improving the trainable parameter requirement from $\mathcal{O}(N^4)$ to $\mathcal{O}(m)$, where $N$ is pixel size and $m$ is number of fibre modes. Finally, this model generalises to new images, outside of the set of training data classes, better than previous models.

PDF Details

AIIM Journal 2021 Journal Article

Prediction of weaning from mechanical ventilation using Convolutional Neural Networks

Yan Jia
Chaitanya Kaul
Tom Lawton
Roderick Murray-Smith
Ibrahim Habli

Weaning from mechanical ventilation covers the process of liberating the patient from mechanical support and removing the associated endotracheal tube. The management of weaning from mechanical ventilation comprises a significant proportion of the care of critically ill intubated patients in Intensive Care Units (ICUs). Both prolonged dependence on mechanical ventilation and premature extubation expose patients to an increased risk of complications and increased health care costs. This work aims to develop a decision support model using routinely-recorded patient information to predict extubation readiness. In order to do so, we have deployed Convolutional Neural Networks (CNN) to predict the most appropriate treatment action in the next hour for a given patient state, using historical ICU data extracted from MIMIC-III. The model achieved 86% accuracy and 0. 94 area under the receiver operating characteristic curve (AUC-ROC). We also performed feature importance analysis for the CNN model and interpreted these features using the DeepLIFT method. The results of the feature importance assessment show that the CNN model makes predictions using clinically meaningful and appropriate features. Finally, we implemented counterfactual explanations for the CNN model. This can help clinicians understand what feature changes for a particular patient would lead to a desirable outcome, i. e. readiness to extubate.

Details DOI

ICLR Conference 2021 Conference Paper

Tomographic Auto-Encoder: Unsupervised Bayesian Recovery of Corrupted Data

Francesco Tonolini
Pablo Garcia Moreno
Andreas C. Damianou
Roderick Murray-Smith

We propose a new probabilistic method for unsupervised recovery of corrupted data. Given a large ensemble of degraded samples, our method recovers accurate posteriors of clean values, allowing the exploration of the manifold of possible reconstructed data and hence characterising the underlying uncertainty. In this set-ting, direct application of classical variational methods often gives rise to collapsed densities that do not adequately explore the solution space. Instead, we derive our novel reduced entropy condition approximate inference method that results in rich posteriors. We test our model in a data recovery task under the common setting of missing values and noise, demonstrating superior performance to existing variational methods for imputation and de-noising with different real data sets. We further show higher classification accuracy after imputation, proving the advantage of propagating uncertainty to downstream tasks with our model.

Details

JMLR Journal 2020 Journal Article

Variational Inference for Computational Imaging Inverse Problems

Francesco Tonolini
Jack Radford
Alex Turpin
Daniele Faccio
Roderick Murray-Smith

Machine learning methods for computational imaging require uncertainty estimation to be reliable in real settings. While Bayesian models offer a computationally tractable way of recovering uncertainty, they need large data volumes to be trained, which in imaging applications implicates prohibitively expensive collections with specific imaging instruments. This paper introduces a novel framework to train variational inference for inverse problems exploiting in combination few experimentally collected data, domain expertise and existing image data sets. In such a way, Bayesian machine learning models can solve imaging inverse problems with minimal data collection efforts. Extensive simulated experiments show the advantages of the proposed framework. The approach is then applied to two real experimental optics settings: holographic image reconstruction and imaging through highly scattering media. In both settings, state of the art reconstructions are achieved with little collection of training data. [abs] [ pdf ][ bib ] &copy JMLR 2020. ( edit, beta )

PDF Details

UAI Conference 2019 Conference Paper

Variational Sparse Coding

Francesco Tonolini
Bjørn Sand Jensen
Roderick Murray-Smith

Unsupervised discovery of interpretable features and controllable generation with high-dimensional data are currently major challenges in machine learning, with applications in data visualisation, clustering and artificial data synthesis. We propose a model based on variational auto-encoders (VAEs) in which interpretation is induced through latent space sparsity with a mixture of Spike and Slab distributions as prior. We derive an evidence lower bound for this model and propose a specific training method for recovering disentangled features as sparse elements in latent vectors. In our experiments, we demonstrate superior disentanglement performance to standard VAE approaches when an estimate of the number of true sources of variation is not available and objects display different combinations of attributes. Furthermore, the new model provides unique capabilities, such as recovering feature exploitation, synthesising samples that share attributes with a given input object and controlling both discrete and continuous features upon generation.

Details

NeurIPS Conference 2018 Conference Paper

Deep, complex, invertible networks for inversion of transmission effects in multimode optical fibres

Oisín Moran
Piergiorgio Caramazza
Daniele Faccio
Roderick Murray-Smith

We use complex-weighted, deep networks to invert the effects of multimode optical fibre distortion of a coherent input image. We generated experimental data based on collections of optical fibre responses to greyscale input images generated with coherent light, by measuring only image amplitude (not amplitude and phase as is typical) at the output of \SI{1}{\metre} and \SI{10}{\metre} long, \SI{105}{\micro\metre} diameter multimode fibre. This data is made available as the {\it Optical fibre inverse problem} Benchmark collection. The experimental data is used to train complex-weighted models with a range of regularisation approaches. A {\it unitary regularisation} approach for complex-weighted networks is proposed which performs well in robustly inverting the fibre transmission matrix, which fits well with the physical theory. A key benefit of the unitary constraint is that it allows us to learn a forward unitary model and analytically invert it to solve the inverse problem. We demonstrate this approach, and show how it can improve performance by incorporating knowledge of the phase shift induced by the spatial light modulator.

PDF Details

IROS Conference 2005 Conference Paper

Human-human haptic collaboration in cyclical Fitts' tasks

Sommer Gentry
Eric Feron
Roderick Murray-Smith

Understanding how humans assist each other in haptic interaction teams could lead to improved robotic aids to solo human dextrous manipulation. Inspired by experiments reported in Reed et al. (2004), which suggested two-person haptically interacting teams could achieve a lower movement time (MT) than individuals for discrete aiming movements of specified accuracy, we report that two-person teams (dyads) can also achieve lower MT for cyclical, continuous aiming movements. We propose a model, called endpoint compromise, for how the intended endpoints of both subjects' motion combine during haptic interaction; it predicts a ratio of /spl radic/2 between slopes of MT fits for individuals and dyads. This slope ratio prediction is supported by our data.

Details

ICRA Conference 2005 Conference Paper

Rehabilitation Robot Cell for Multimodal Standing-Up Motion Augmentation

Roman Kamnik
Tadej Bajd
John Williamson 0001
Roderick Murray-Smith

The paper presents a robot cell for multimodal standing-up motion augmentation. The robot cell is aimed at augmenting the standing-up capabilities of impaired or paraplegic subjects. The setup incorporates the rehabilitation robot device, functional electrical stimulation system, measurement instrumentation and cognitive feedback system. For controlling the standing-up process a novel approach was developed integrating the voluntary activity of a person in the control scheme of the rehabilitation robot. The simulation results demonstrate the possibility of “patient-driven” robot-assisted standing-up training. Moreover, to extend the system capabilities, the audio cognitive feedback is aimed to guide the subject throughout rising. For the feedback generation a granular synthesis method is utilized displaying high-dimensional, dynamic data. The principle of operation and example sonification in standing-up are presented. In this manner, by integrating the cognitive feedback and “patient-driven” actuation systems, an effective motion augmentation system is proposed in which the motion coordination is under the voluntary control of the user.

Details

NeurIPS Conference 2002 Conference Paper

Gaussian Process Priors with Uncertain Inputs Application to Multiple-Step Ahead Time Series Forecasting

Agathe Girard
Carl Rasmussen
Joaquin Candela
Roderick Murray-Smith

We consider the problem of multi-step ahead prediction in time series analysis using the non-parametric Gaussian process model. -step ahead forecasting of a discrete-time non-linear dynamic system can be per- formed by doing repeated one-step ahead predictions. For a state-space at time model of the form is based on the point estimates of the previous outputs. In this pa- per, we show how, using an analytical Gaussian approximation, we can formally incorporate the uncertainty about intermediate regressor values, thus updating the uncertainty on the current prediction.

PDF Details

NeurIPS Conference 1998 Conference Paper

Robot Docking Using Mixtures of Gaussians

Matthew Williamson
Roderick Murray-Smith
Volker Hansen

This paper applies the Mixture of Gaussians probabilistic model, com(cid: 173) bined with Expectation Maximization optimization to the task of sum(cid: 173) marizing three dimensional range data for a mobile robot. This provides a flexible way of dealing with uncertainties in sensor information, and al(cid: 173) lows the introduction of prior knowledge into low-level perception mod(cid: 173) ules. Problems with the basic approach were solved in several ways: the mixture of Gaussians was reparameterized to reflect the types of objects expected in the scene, and priors on model parameters were included in the optimization process. Both approaches force the optimization to find 'interesting' objects, given the sensor and object characteristics. A higher level classifier was used to interpret the results provided by the model, and to reject spurious solutions.

PDF Details