The Compass CDT will be well-represented at this year’s International Conference on Machine Learning (ICML), with several of its students having papers accepted by the prestigious event.

One of these has been selected for oral presentation and one as a ‘spotlight position paper’, while another was co-authored by more than one Compass student – indicative of the strength and depth of the Centre’s machine learning and AI research:

• Oral presentation: ‘Score Matching with Missing Data’
  Josh Givens (lead author); Song Liu; Henry W J Reeve

• Spotlight: ‘Don’t Use the CLT in LLM Evals With Fewer Than a Few Hundred Datapoints’
  Sam Bowyer (lead author); Laurence Aitchison; Desi R. Ivanova

• Poster presentation: ‘Function-Space Learning Rates‘
  Ed Milsom (lead author) and Ben Anson (co-author); Laurence Aitchison

• Poster presentation: ‘Flexible Tails for Normalizing Flows’
  Tennessee Hickling (lead author); Dennis Prangle

This strong presence at ICML, which takes place in Vancouver next week, forms part of a packed conference summer, with Compass students presenting at other important UK and international events, including UAI 2025 in Brazil, and the ERSA Congress in Greece. (See full list).

ICML 2025 will take place in Vancouver, Canada from 13 to 19 July 2025.

“It’s fantastic to see our students making valuable contributions to one of the leading conferences in this field, and especially rewarding to have an oral presentation and spotlight paper as part of that,” said Professor Nick Whiteley, Director of the Compass CDT.

“Alongside NeurIPS and ICLR, both of which Compass students and alumni have contributed to in recent years, ICML provides an opportunity to showcase our work to machine learning researchers and practitioners from across academia and industry.

We are proud of the cutting-edge projects being undertaken by students and colleagues in the Centre, and across the wider Institute for Statistical Science here in the School of Mathematics, which span data science, statistics, machine learning and AI.

We’ve had some positive outcomes already this year, with our first Compass graduates finding employment with a range of organisations in the private and public sectors, as well as in academia, and current students securing some exciting internships.

Having papers accepted at high impact conferences, including ICML, and being involved in well-regarded events at home and overseas, is a fantastic way to round off the academic year.”

With Compass student Sherman Khoo winning an Early Career Researcher Poster Award in June, at Bayes Comp 2025 – a meeting of the Bayesian Computation Section of the International Society for Bayesian Analysis –  the conference season is already off to a successful start.

Compass CDT student Sherman Khoo (furthest left) being announced as a winner of an Early Career Researcher Poster Award at Bayes Comp 2025, held in Singapore in June.

The fact that some papers accepted at conferences this year, including one at ICML, were co-authored by more than one Compass student has been particularly rewarding, as one of the Centre’s aims is to foster collaboration, and to build a community of researchers.

Emerald Dilworth will present a poster at Uncertainty in Artificial Intelligence (UAI), summarising a paper for which she was lead author, and Compass graduate Ed Davis was one of the co-authors, as was Compass Co-director Professor Daniel Lawson. “I reached out to Ed with my initial ideas, and we grew them together,” says Emerald.

“Some of the work belongs to each of us as parts we contributed individually, and other parts we worked on together, which is something we enjoyed. It’s a valuable, positive experience to collaborate on research, when a lot of what we do as PhD students can be solo work.”

This collaborative culture, alongside awards, presentations, spotlight slots, and a strong presence at ICML and other events, makes for a positive 2025 conference season, and suggests this trend will continue in future years, as students and alumni continue to gain experience.

“Travelling to Vancouver in July, and representing the University of Bristol, and the UK research community, at a forum as important as ICML, is a fantastic opportunity,” says Professor Whiteley. “The insights students gain will help to shape their future research.

The fact that students from both the 2021 and 2022 Compass cohorts will contribute to that conference, while all cohorts will be represented at a diverse range of leading events this summer, shows the breadth of Compass expertise. The research our students showcase, the knowledge they gain, and the networks they build, will have benefits for a long time to come.”

Compass student Cecina Babich Morrow (third from right) was a panellist at a GW4 AI and Data Science event focused on climate and health, held in Bristol in June.

 

Compass at ICML 2025

July 2025 – Vancouver, Canada

Student: Josh Givens (lead author)
Co-authors: Song Liu; Henry W J Reeve
Contribution: Oral presentation
Overview: ‘Score Matching with Missing Data’. Score matching is a technique used to learn intractable data distributions and serves as a foundational component of state-of-the-art image generation methods. In this paper, we adapt score matching (and its various extensions) to enable the learning of the original data distribution using only corrupted samples, where parts of each observation are missing or contain NaN values.

Student: Sam Bowyer (lead author)
Co-authors: Laurence Aitchison; Desi R. Ivanova
Contribution: Spotlight Position Paper
Overview: ‘Position: Don’t Use the CLT in LLM Evals With Fewer Than a Few Hundred Datapoints’. We argue that AI researchers need to improve the techniques they use to calculate error bars (i.e. confidence/credible intervals) over model-performance on benchmark test-sets (‘evals’). We examine the failure of common approaches that rely on the Central Limit Theorem (CLT) and suggest Bayesian and frequentist alternatives in a variety of eval settings, such as questions organised in subtasks/clusters, and comparisons between two models.

Student: Ed Milsom (lead author) and Ben Anson (co-author)
Co-author: Laurence Aitchison
Contribution: Poster
Overview: ‘Function-Space Learning Rates’. This will help AI researchers better understand why current AI systems work well and how to improve them in the future. We utilise our method to tune very large AI systems by tuning a small, cheap model and then copying the settings to the large model (which would be very computationally expensive to tune by itself). This is not usually possible, because the optimal settings change between small and large AI systems, but with our method, we can predict and therefore correct these changes.

Student: Tennessee Hickling (lead author)
Co-authors Dennis Prangle
Contribution: Poster
Overview: ‘Flexible Tails for Normalizing Flows’. Modern machine learning methods model uncertainty by transforming simple random inputs into data-like outputs, but they often underestimate extreme events. While some recent approaches inject extreme inputs directly, this can cause models to behave poorly. We propose a different solution: adding a final step that enables models to generate extreme outcomes from non-extreme inputs, improving performance on data with heavy tails.

‘Paired question model comparison setting’ – from Sam Bowyer’s ICML Spotlight Position Paper.

 

Compass at UAI 2025

July 2025 – Rio de Janeiro, Brazil

Student: Emerald Dilworth (lead author) and Ed Davis (co-author)
Co-authors: Daniel Lawson (Compass Co-director)
Contribution: Poster
Overview: ‘Valid Bootstraps for Network Embeddings with Applications to Network Visualisation’. Quantifying uncertainty in networks is an important step in modelling relationships and interactions between entities. Under certain assumptions of the network, we utilise an exchangeable network test that can empirically validate bootstrap samples generated by any method, by considering if embeddings of the observed and bootstrapped network that are statistically indistinguishable. Existing methods fail this test, so we propose a principled, distribution-free network bootstrap using k-nearest neighbour smoothing, which can pass this exchangeable network test in many synthetic and real-data scenarios. We demonstrate the utility of this work in combination with the popular data visualisation method t-SNE, where uncertainty estimates from bootstrapping are used to explain whether visible structures represent real statistically sound structures.

Student: Emma Ceccherini (lead author)
Co-authors: Ian Gallagher; Andrew Jones; Daniel Lawson (Compass Co-director)
Contribution: Poster
Overview: ‘Unsupervised Attributed Dynamic Network Embedding with Stability Guarantees’. Most AI/ML methods have exchangeability or stability properties, loosely speaking, for the same input they produce the same output up to noise, however, this is not always true for dynamic graph embedding. We propose a dynamic attributed embedding that collectively embeds attributes and network information in the same low-dimensional space, and we prove uniform convergence, which establishes stability. As a result, our method performs better than competitors on downstream tasks.

A simulated results figure from Emerald Dilworth’s paper, co-authored with Ed Davis.

 

Compass at GOFCP 2025

(Goodness-of-fit, Change-point and Related Problems)

August 2025 – Charles University, Prague, Czech Republic

Student: Dylan Dijk
Contribution: Poster
Overview: This poster presents work on generalising the assumptions behind a widely used econometric model (the model is partly described in Dylan’s blog from July 2024). In real-world applications – especially in finance – large datasets often contain extreme values, particularly during periods of crisis. The goal is to adapt the model’s theoretical foundations to ensure reliable performance even when such heavy-tailed data are present.

Student: Yuqi Zhang
Contribution: Poster
Overview: Presents a method developed jointly with Dr Haeran Cho (Compass Projects Coordinator). We introduce a multiscale, bandwidth-free procedure for detecting multiple change points in large approximate factor models. The method combines the Narrowest-over-Threshold (NOT) principle with Seeded Binary Segmentation (SBS) to efficiently identify structural changes in high-dimensional time series without requiring prior bandwidth selection. The poster presents a method developed jointly with Dr Haeran Cho, designed for multiple change points in high-dimensional factor models.

 

Compass contributions to other events:

Random Networks Workshop
Date/location: May 2025 – University of Sheffield, UK
Student: Ollie Baker (lead author)
Co-authors: Carl P. Dettmann
Contribution: Contributed talk
Overview: ‘Entropy of Random Geometric Graphs in High and Low Dimensions’. The paper uses information theory to discuss whether or not we can detect the geometric embedding of a spatial network when the dimension of the embedding gets very large. We then use this to derive a bound on the amount of information saved by embedding a network in a lower dimensional space.

Bayes Comp 2025
Date/location: June 2025 – National University of Singapore
Student: Sherman Khoo (lead author)
Contribution: Poster
Overview: ‘Approximate Maximum Likelihood Estimation with Local Score Matching’. We study the problem of likelihood maximization when the likelihood function is intractable, but model simulations are readily available. We propose a sequential, gradient-based optimization method that directly models the Fisher score based on a local score matching technique which uses simulations from a localized region around each parameter iterate.

UK AI Conference 2025
Date/location: June 2025 – London, UK
Student: Rachel Wood (lead author)
Contribution: Poster
Overview: We first consider how to define an anomaly in a network, then explore using existing reliable and efficient embedding methods for this purpose. Most methods are only present for the correct (unknown) latent dimension $d$ and performance may deteriorate for other choices of $d$, so we propose an approach which retains accuracy when the dimension choice is mis-specified, offering a more robust solution for anomaly detection in dynamic and uncertain network environments.

GW4 AI and Data Science: AI, Climate and Health
Date/location: June 2025 – University of Bristol, UK
Student: Cecina Babich Morrow
Contribution: Lightning talk, panel member and poster
Overview: ‘From risk to action: Climate decision-making under deep uncertainty’. How can we make robust climate adaptation decisions despite our uncertainty about both the level of climate risk and the characteristics of our adaptations? I showed how uncertainty and sensitivity analysis might be helpful in addressing these issues.

46^th Annual Conference of the International Society for Clinical Biostatistics (ISCB)
Date/location: August 2025 – University of Basel, Switzerland
Student: Vera Hudak (lead author)
Co-authors: Hayley Jones; Nicky J. Welton; Efthymia Derezea
Contribution: Poster
Overview: ‘Evaluating Diagnostic Tests Against Composite Reference Standards: Quantifying and Adjusting for Bias’. Our research focuses on diagnostic test accuracy studies where gold standard testing is only performed on a subset of study participants, specifically those with certain results from some initial imperfect reference test. We have quantified the bias that can arise in these scenarios and proposed a method to adjust for it, which was evaluated using a simulation study.

64^th European Regional Science Association (ERSA) Congress: ‘Regional Science in Turbulent Times. In search of a resilient, sustainable and inclusive future’
Date/location: August 2025 – Panteion University, Athens, Greece
Student: Emerald Dilworth (presenter)
Co-authors: Emmanouil Tranos; Daniel Lawson (Compass Co-director)
Contribution: Special session presentation
Overview: ‘The Twin Transition in the UK through LLM labelled Web Crawl Data’. Many existing tools for tracking green and digital industries are expensive, limited in scale, or updated too infrequently, making it difficult to design effective policies for technological progress. To address this, we use freely available web data to identify UK firms involved in green and digital technologies by fine-tuning a language model to classify company websites. This allows us to track how these industries have evolved across regions and over time from 2014 to 2024, at a yearly time scale.

Global Optimization Workshop 2025
Date/location: September 2025 – KTH Royal Institute of Technology, Stockholm, Sweden
Student: Ettore Fincato (co-author)
Co-authors: Christophe Andrieu; Nicolas Chopin; Mathieu Gerber
Contribution: Paper presentation
Overview: ‘Gradient-free optimization via integration’. We develop and analyse a novel approach to optimize functions not assumed to be convex, differentiable or even continuous. The idea is to fit recursively the objective function to a parametric family of distributions, using a Bayesian update followed by a reprojection back onto the chosen family. Practically, the approach enables the optimization of a broad class of objectives via Monte Carlo sampling; theoretically, it establishes a link with gradient-based methods with smoothing.

Exact entropy curves from Ollie Baker’s paper, presented at the Random Networks Workshop.

 

A post by Daniel Gardner, PhD student on the Compass programme.

Introduction

My research focuses on genetic interaction networks within lung cancer cells. Our (long-term) aim is to model such networks dynamically using a Boolean modelling framework, and then use this to tie changes in cancer cells’ physiology to certain, often mutated, genes of interest.

Aims and problems

This blog post will focus on the challenge we are currently working on: constructing the model itself. This is often the most challenging element of the research, as it underpins all results going forward, and often there does not exist enough data to fully define a unique model.

In some respects this is acceptable, as Boolean modelling is more of a qualitative approach. Each node in the network is a ‘species’, be that a gene, protein, small molecule, etc. Each directed, labelled edge is either ‘activating’, if an increase in species A causes an increase in species B, or ‘inhibiting’, if the opposite is the case [1].

With this definition, a lot of papers we have looked at define their model purely from the literature [2], [3], [4], either manually mining links, or using pre-existing databases like the Kyoto Encyclopedia of Genes and Genomes (KEGG)[6].

What we are more interested in are methods deriving these models in a far more quantitative way, straight from transcriptomic data. Whilst some of the papers referenced above justify their hand-built models in retrospect by showing they can replicate real-world results [3], we wish to work the other way round – beginning at the real-world results and then using a reverse engineering approach.

Figure 1: The Boolean model used in [2], based off a similar model constructed in [4]. It contains 98 nodes (species) and 254 directed edges (labelled interactions).

Potential solutions

The solutions we have found can be broadly split into two categories: methods that go from:

Raw Data → Interaction Network

and similarly:

Interaction Network → Boolean Model

The former is a much more difficult challenge. Generally, in a published network, each edge will reference experimental work that justifies, e.g. ‘A activates B’. However, data-frames which contain many cell-line perturbation experiments in one are hard to come by, and expensive to perform [5]. The problem is often also undetermined since the solution-space for a potential network is far greater than the amount of data available. One option we may look into in the future, however, involves using other modelling techniques, such as ODEs or Bayesian networks.

The challenge of reverse engineering a Boolean network from a pre-built network is much more feasible. The main problem in this case is considering complex interactions. For example, if we had ’A inhibits C’ and ’B activates C’, how do they work in tandem?

Figure 2: Part of the optimisation algorithm from [7] applied to a toy model. In D, we classify each species in the network. All non-compressed nodes are those which we have data to train on. In E, we construct the hypergraph, where for any pair of combined interactions, both the ‘AND’ and ‘OR’ case are considered.

Sticking to the Boolean framework, these two interactions can either be joined through an ‘AND’ relation, or an ‘OR’ relation. For several proteins affecting one specific protein, the combinations of Boolean rules are non-trivial.

One paper we found that deals with this problem well is Saez-Rodriguez et al. [7], which attempts to train a hypergraph of the interaction network to cell line assay data. It contains a number of different techniques to do with graph and state space reduction, as well as some heuristic rules on which complex interactions to target. For example, it is unlikely in biology for a protein to require multiple other species to necessitate a change in function, so we can remove ‘AND’ links of more than N complex interactions from the state space.

One other model component we are looking for, which we have not currently looked into properly, is a ‘layered’ model, which includes different levels of genomic interaction. For example, many papers we have read use ‘protein interaction network (PIN)’ and ‘gene regulatory network (GRN)’ interchangeably. Whilst the two are greatly related, drawing a one-to-one equivalence between the two in all cases is incorrect.

Conclusion and future plans

Starting directly from data to build a network is perhaps too ambitious a challenge, especially with the limited data available. In fact, even to train a Boolean network for optimisation requires quite specific cell-line perturbation data. It could be that we make do with a network partially trained on limited data, and the rest taken from prior knowledge in the literature.

One promising sign is that [7] finds that it is best to begin with ’too many’ interactions in a literature-curated interaction network, and then ’prune’ spurious interactions via network optimisation. This is due to these large networks being built from many different sources, some using different tissue, conditions, etc. Therefore, when we desire a model specific to lung adenocarcinoma data, it is natural for the training to remove many of these genetic interactions.

In the future, we aim for this research topic to simply be one section of the wider project. Once we decide upon the most justified Boolean model for lung cancer, we aim to use patient mRNA and mutation data to personalise the models, in order to predict patient specific cell phenotype probabilities. Using this, along with multi-layer protein imaging data from Cancer Research UK, we aim to find a statistically significant link between certain gene mutations, and the resulting shape and, therefore, phenotype of a tumour of cancer cells.

Thank you for reading this blog post. If you have any questions, please feel free to get in touch with me at: daniel.gardner@bristol.ac.uk

References

[1] Abou-Jaoudé, W., Traynard, P., Monteiro, P. T., Saez- Rodrıguez, J., Helikar, T., Thieffry, D., and Chaouiya, C. (2016). Logical modeling and dynamical analysis of cellular networks. Frontiers in Genetics, 7.

[2] Béal, J., Montagud, A., Traynard, P., Barillot, E., and Calzone, L. (2019). Personalization of logical models with multi-omics data allows clinical stratification of patients. Frontiers in Physiology, 9.

[3] Cohen, D. P. A., Martignetti, L., Robine, S., Barillot, E., Zinovyev, A., and Calzone, L. (2015). Mathematical modelling of molecular pathways enabling tumour cell invasion and migration. PLOS Computational Biology, 11.

[4] Fumiã, H. (2013). Boolean network model for cancer pathways: Predicting carcinogenesis and targeted therapy outcomes. PloS one, 8:e69008.

[5] Galindez, G., Sadegh, S., Baumbach, J., Kacprowski, T., and List, M. (2023). Network-based approaches for modeling disease regulation and progression. Computational and Structural Biotechnology Journal, 21:780–795. 4

[6] Kanehisa, M. and Goto, S. (2000). Kegg: Kyoto encyclopedia of genes and genomes. Nucleic Acids Research, 28(1):27–30.

[7] Saez-Rodriguez, J., Alexopoulos, L. G., Epperlein, J., Samaga, R., Lauffenburger, D. A., Klamt, S., and Sorger, P. K. (2009). Discrete logic modelling as a means to link protein signalling networks with functional analysis of mammalian signal transduction. Molecular Systems Biology, 5(1):331.