Student perspectives: Compass Annual Conference 2024

A post by Compass students Ben Anson, Ollie Baker, Codie Wood and Rachel Wood.

Introduction

This October, we held our third annual Compass Conference. Unlike previous years, when the event was held in the University’s Fry Building, this time it took place at M Shed, offering scenic views of Bristol harbour. It was a great day for past and present Compass students, academics, and industry partners to come together and discuss this year’s theme: “The Future of Data Science”. With recent advances in machine learning and AI, it felt like a fitting time to learn from each other’s perspectives and to share ideas about how to move forward in this exciting space.

Panoramic view of Bristol harbour, as seen from M Shed

Student Research Talks

The morning started with four ten-minute research talks from Compass students. First was Rahil Morjaria‘s talk on “Group Testing” which explored current developments in the field, including algorithms and information-theoretic limits.

Following this, Kieran Morris presented “A Trip to Bregman Geometry and Applications”, considering advancements such as natural gradient methods, Bregman K-means clustering, and EM-projection algorithms that Bregman Geometry has enabled.

Ettore Fincato talked us through “Gradient-Free Optimisation via Integration”, focusing on a novel yet easy-to-implement algorithm for optimisation using Monte Carlo methods. Finally, Ed Milsom spoke about “Data Modalities and the Bias-Variance Decomposition”, taking us through a history of neural networks and speculating about why certain data types are so powerful, and why the future of general-purpose AI must be multi-modal.

Student Lightning Talks

The lightning talks challenged ten students to present on a topic in just three minutes. The ability to quickly convey a message in an engaging and understandable manner, to an audience with diverse backgrounds, is crucial in both academia and industry, and the students rose to the occasion.

Their talks captured the interest of the audience and inspired interesting questions that forced the students to think on their feet. Topics ranged from neural networks and large language models (LLMs), to making music using mathematics.

Compass Alumni Panel

This year’s conference panel, chaired by Compass CDT Director, Professor Nick Whiteley, offered an engaging look into the professional journeys of Compass alumni Dominic Owens, Jake Spiteri and Michael Whitehouse since  completing their PhDs. With shared experiences in finance, each panelist provided unique insights into the early career landscape and the skills that helped them succeed.

Jake delved into the details of his day-to-day work in the financial sector, while Dominic discussed the challenge and dedication required to secure a role through extensive networking and job applications. Michael shared details of his transition from finance to epidemiological research. Together, they sparked valuable discussions on what the future of data science might hold for upcoming Compass graduates.

Special Guest Lecture

The conference concluded with an enlightening special guest lecture by Professor Aline Villavicencio, Director of the Institute for Data Science and Artificial Intelligence at the University of Exeter. Her talk, “Testing the Idiomatic Language Limits of Foundation Models: The Strange Case of the Idiomatic Eager Beaver in Cloud Nine,” offered a fascinating counterpoint to the current enthusiasm surrounding LLMs.

Drawing from her research in Natural Language Processing (NLP), Professor Villavicencio demonstrated how even today’s most advanced models struggle with aspects of language that humans master naturally – particularly idioms and multi-word expressions. She illustrated a persistent gap between machine and human linguistic capabilities, reminding us that the path to truly human-like language understanding remains long and complex.

She also shared her perspective on the cyclical nature of NLP research, noting how, throughout her career, there have been multiple predictions about NLP research becoming obsolete as models improve. Yet, as her work on datasets like SemEval (Semantic Evaluation) shows, there remain fundamental challenges in representing and understanding idiomatic language.

Concluding remarks

The successful day of talks, poster sessions and networking culminated with Professor Whiteley sharing his thoughts on what we learned throughout the event. He concluded that the future of our field is certain to be exciting and will encompass a huge range of different areas and ideas. This year’s conference embodied this by providing a platform for students, academics, and industry professionals to share new insights from many different sectors, and to form strong relationships to help forge a path to the future of data science.

 

Past conferences

Student Perspectives: SPREE Methods for Small Area Estimation

A post by Codie Wood, PhD student on the Compass programme.

This blog post is an introduction to structure preserving estimation (SPREE) methods. These methods form the foundation of my current work with the Office for National Statistics (ONS), where I am undertaking a six-month internship as part of my PhD. During this internship, I am focusing on the use of SPREE to provide small area estimates of population characteristic counts and proportions.

Small area estimation

Small area estimation (SAE) refers to the collection of methods used to produce accurate and precise population characteristic estimates for small population domains. Examples of domains may include low-level geographical areas, or population subgroups. An example of an SAE problem would be estimating the national population breakdown in small geographical areas by ethnic group [2015_Luna].

Demographic surveys with a large enough scale to provide high-quality direct estimates at a fine-grain level are often expensive to conduct, and so smaller sample surveys are often conducted instead.

SAE methods work by drawing information from different data sources and similar population domains in order to obtain accurate and precise model-based estimates where sample counts are too small for high quality direct estimates. We use the term small area to refer to domains where we have little or no data available in our sample survey.

SAE methods are frequently relied upon for population characteristic estimation, particularly as there is an increasing demand for information about local populations in order to ensure correct allocation of resources and services across the nation.

Structure preserving estimation

Structure preserving estimation (SPREE) is one of the tools used within SAE to provide population composition estimates. We use the term composition here to refer to a population break down into a two-way contingency table containing positive count values. Here, we focus on the case where we have a population broken down into geographical areas (e.g. local authority) and some subgroup or category (e.g. ethnic group or age).

Orginal SPREE-type estimators, as proposed in [1980_Purcell], can be used in the case when we have a proxy data source for our target composition, containing information for the same set of areas and categories but that may not entirely accurately represent the variable of interest. This is usually because the data are outdated or have a slightly different variable definition than the target.

We also incorporate benchmark estimates of the row and column totals for our composition of interest, taken from trusted, quality assured data sources and treated as known values. This ensures consistency with higher level known population estimates. SPREE then adjusts the proxy data to the estimates of the row and column totals to obtain the improved estimate of the target composition.

IMG_1633

An illustration of the data required to produce SPREE-type estimates.

In an extension of SPREE, known as generalised SPREE (GSPREE) [2004_Zhang], the proxy data can also be supplemented by sample survey data to generate estimates that are less subject to bias and uncertainty than it would be possible to generate from each source individually. The survey data used is assumed to be a valid measure of the target variable (i.e. it has the same definition and is not out of date), but due to small sample sizes may have a degree of uncertainty or bias for some cells.

The GSPREE method establishes a relationship between the proxy data and the survey data, with this relationship being used to adjust the proxy compositions towards the survey data.

IMG_1634 (1)

An illustration of the data required to produce GSPREE estimates.

GSPREE is not the only extension to SPREE-type methods, but those are beyond the scope of this post. Further extensions such as Multivariate SPREE are discussed in detail in [2016_Luna].

Original SPREE methods

First, we describe original SPREE-type estimators. For these estimators, we require only well-established estimates of the margins of our target composition.

We will denote the target composition of interest by $\mathbf{Y} = (Y{aj})$, where $Y{aj}$ is the cell count for small area $a = 1,\dots,A$ and group $j = 1,\dots,J$. We can write $\mathbf Y$ in the form of a saturated log-linear model as the sum of four terms,

$$ \log Y_{aj} = \alpha_0^Y + \alpha_a^Y + \alpha_j^Y + \alpha_{aj}^Y.$$

There are multiple ways to write this parameterisation, and here we use the centered constraints parameterisation given by $$\alpha_0^Y = \frac{1}{AJ}\sum_a\sum_j\log Y_{aj},$$ $$\alpha_a^Y = \frac{1}{J}\sum_j\log Y_{aj} – \alpha_0^Y,$$ $$\alpha_j^Y = \frac{1}{A}\sum_a\log Y_{aj} – \alpha_0^Y,$$ $$\alpha_{aj}^Y = \log Y_{aj} – \alpha_0^Y – \alpha_a^Y – \alpha_j^Y,$$

which satisfy the constraints $\sum_a \alpha_a^Y = \sum_j \alpha_j^Y = \sum_a \alpha_{aj}^Y = \sum_j \alpha_{aj}^Y = 0.$

Using this expression, we can decompose $\mathbf Y$ into two structures:

  1. The association structure, consisting of the set of $AJ$ interaction terms $\alpha_{aj}^Y$ for $a = 1,\dots,A$ and $j = 1,\dots,J$. This determines the relationship between the rows (areas) and columns (groups).
  2. The allocation structure, consisting of the sets of terms $\alpha_0^Y, \alpha_a^Y,$ and $\alpha_j^Y$ for $a = 1,\dots,A$ and $j = 1,\dots,J$. This determines the size of the composition, and differences between the sets of rows (areas) and columns (groups).

Suppose we have a proxy composition $\mathbf X$ of the same dimensions as $\mathbf Y$, and we have the sets of row and column margins of $\mathbf Y$ denoted by $\mathbf Y_{a+} = (Y_{1+}, \dots, Y_{A+})$ and $\mathbf Y_{+j} = (Y_{+1}, \dots, Y_{+J})$, where $+$ substitutes the index being summed over.

We can then use iterative proportional fitting (IPF) to produce an estimate $\widehat{\mathbf Y}$ of $\mathbf Y$ that preserves the association structure observed in the proxy composition $\mathbf X$. The IPF procedure is as follows:

  1. Rescale the rows of $\mathbf X$ as $$ \widehat{Y}_{aj}^{(1)} = X_{aj} \frac{Y_{+j}}{X_{+j}},$$
  2. Rescale the columns of $\widehat{\mathbf Y}^{(1)}$ as $$ \widehat{Y}_{aj}^{(2)} = \widehat{Y}_{aj}^{(1)} \frac{Y_{a+}}{\widehat{Y}_{a+}^{(1)}},$$
  3. Rescale the rows of $\widehat{\mathbf Y}^{(2)}$ as $$ \widehat{Y}_{aj}^{(3)} = \widehat{Y}_{aj}^{(2)} \frac{Y_{+j}}{\widehat{Y}_{+j}^{(2)}}.$$

Steps 2 and 3 are then repeated until convergence occurs, and we have a final composition estimate denoted by $\widehat{\mathbf Y}^S$ which has the same association structure as our proxy composition, i.e. we have $\alpha_{aj}^X = \alpha_{aj}^Y$ for all $a \in \{1,\dots,A\}$ and $j \in \{1,\dots,J\}.$ This is a key assumption of the SPREE implementation, which in practise is often restrictive, motivating a generalisation of the method.

Generalised SPREE methods

If we can no longer assume that the proxy composition and target compositions have the same association structure, we instead use the GSPREE method first introduced in [2004_Zhang], and incorporate survey data into our estimation process.

The GSPREE method relaxes the assumption that $\alpha_{aj}^X = \alpha_{aj}^Y$ for all $a \in \{1,\dots,A\}$ and $j \in \{1,\dots,J\},$ instead imposing the structural assumption $\alpha_{aj}^Y = \beta \alpha_{aj}^X$, i.e. the association structure of the proxy and target compositions are proportional to one another. As such, we note that SPREE is a particular case of GSPREE where $\beta = 1$.

Continuing with our notation from the previous section, we proceed to estimate $\beta$ by modelling the relationship between our target and proxy compositions as a generalised linear structural model (GLSM) given by
$$\tau_{aj}^Y = \lambda_j + \beta \tau_{aj}^X,$$ with $\sum_j \lambda_j = 0$, and where $$ \begin{align} \tau_{aj}^Y &= \log Y_{aj} – \frac{1}{J}\sum_j\log Y_{aj},\\
&= \alpha_{aj}^Y + \alpha_j^Y,
\end{align}$$ and analogously for $\mathbf X$.

It is shown in [2016_Luna] that fitting this model is equivalent to fitting a Poisson generalised linear model to our cell counts, with a $\log$ link function. We use the association structure of our proxy data, as well as categorical variables representing the area and group of the cell, as our covariates. Then we have a model given by $$\log Y_{aj} = \gamma_a + \tilde{\lambda}_j + \tilde{\beta}\alpha_{aj}^X,$$ with $\gamma_a = \alpha_0^Y + \alpha_a^Y$, $\tilde\lambda_j = \alpha_j^Y$ and $\tilde\beta \alpha_{aj}^X = \alpha_{aj}^Y.$

When fitting the model we use survey data $\tilde{\mathbf Y}$ as our response variable, and are then able to obtain a set of unbenchmarked estimates of our target composition. The GSPREE method then benchmarks these to estimates of the row and column totals, following a procedure analagous to that undertaken in the orginal SPREE methodology, to provide a final set of estimates for our target composition.

ONS applications

The ONS has used GSPREE to provide population ethnicity composition estimates in intercensal years, where the detailed population estimates resulting from the census are outdated [2015_Luna]. In this case, the census data is considered the proxy data source. More recent works have also used GSPREE to estimate counts of households and dwellings in each tenure at the subnational level during intercensal years [2023_ONS].

My work with the ONS has focussed on extending the current workflows and systems in place to implement these methods in a reproducible manner, allowing them to be applied to a wider variety of scenarios with differing data availability.

References

[1980_Purcell] Purcell, Noel J., and Leslie Kish. 1980. ‘Postcensal Estimates for Local Areas (Or Domains)’. International Statistical Review / Revue Internationale de Statistique 48 (1): 3–18. https://doi.org/10/b96g3g.

[2004_Zhang] Zhang, Li-Chun, and Raymond L. Chambers. 2004. ‘Small Area Estimates for Cross-Classifications’. Journal of the Royal Statistical Society Series B: Statistical Methodology 66 (2): 479–96. https://doi.org/10/fq2ftt.

[2015_Luna] Luna Hernández, Ángela, Li-Chun Zhang, Alison Whitworth, and Kirsten Piller. 2015. ‘Small Area Estimates of the Population Distribution by Ethnic Group in England: A Proposal Using Structure Preserving Estimators’. Statistics in Transition New Series and Survey Methodology 16 (December). https://doi.org/10/gs49kq.

[2016_Luna] Luna Hernández, Ángela. 2016. ‘Multivariate Structure Preserving Estimation for Population Compositions’. PhD thesis, University of Southampton, School of Social Sciences. https://eprints.soton.ac.uk/404689/.

[2023_ONS] Office for National Statistics (ONS), released 17 May 2023, ONS website, article, Tenure estimates for households and dwellings, England: GSPREE compared with Census 2021 data

Compass student publishes article in Frontiers

Compass student Dan Milner and his academic supervisors have published an article in Frontiers, one of the most cited and largest research publishers in the world. Dan’s work is funded in collaboration with ILRI (International Livestock Research Institute). (more…)

ILRI sponsors Compass PhD project 

We are excited to announce a new partnership between Compass – the EPSRC Centre for Doctoral Training in Computational Statistics and Data Science – and the International Livestock Research Institute (ILRI).

International Livestock Research Institute

The first step in this new partnership is a co-funded and co-created PhD research project entitled A spatially explicit assessment of agro-pastoral sustainability in Kenya and Ethiopia. The aim of the PhD project is to develop a framework for the assessment of sustainability dynamics in ecologically important areas used by agro-pastoral and pastoral households. Mountainous areas are important water towers and reserves of biodiversity in East Africa, and conservation of such areas is important to stop degradation of the surrounding arid lowlands. However, population pressure and food demands continue to rise, so a sustainable balance between land use and land stewardship must be struck. The PhD project will build upon methods of agricultural sustainability assessment, and make use of spatial statistics to bring together data from household surveys, soil and water measurements, and remote sensing. The resulting analysis will contribute to the understanding of current human-environment interactions in the two study locations, and form the basis for developing scenarios considering the pros and cons of potential future changes. The PhD contributes to the ESSA project, and will operate in Yabelo, South-East Ethiopia, and the Taita Hills, South East Kenya.

“Coming from a geography background, the Compass-ILRI partnership is a fantastic opportunity for me to elevate my skill-set and apply cutting edge statistical techniques to the challenge of sustainable food security. ILRI are a world leader in agricultural research and I am really looking forward to learning from them and contributing to their important goal.” Dan Milner, Compass PhD student.

Dan Milner, Compass-ILRI PhD student

The International Livestock Research Institute (ILRI) works for better lives through livestock in developing countries. ILRI is co-hosted by Kenya and Ethiopia, has 14 offices across Asia and Africa, employs some 700 staff.

(more…)

Skip to toolbar