post by Daniel Swann (2021 cohort)

In early June 2022, I attended a PhD summer school entitled ‘Inquiry through making and playing with objects, spaces and situations.’ The summer school lasted for three days and was organised by Design School Kolding, although was hosted by Aalborg University at their Copenhagen campus. I was looking forward to attending the summer school: I had identified play as an important theme to my research and was also engaging with theoretical approaches to design, particularly in the way that it can inform interaction. The summer school was not specifically focused on children and young people, but I felt this would be a positive as it may allow me to reengage with my research from a different perspective—which I understood as one of the main benefits of attending a summer school as a PhD student.

On the way to Copenhagen, I went over the provided list of reading resources which were curiously broad in scope. It included seminal texts from design theory, research papers from the field of Human-Computer Interaction, experimental autoethnography, and examples of futures studies (an area which was almost entirely new to me). As well as providing the reading list in advance, the organisers asked that each participant brought an object or image that somehow embodied the focus of our research. This provided the basis for introductory conversations with the group on day one of the summer school.

After we had done our introductions, the organisers each made a short presentation that established some of the key themes and ideas of the summer school. This included the many ways we can think about play, especially in the context of making, and a broad overview of design anthropology. These presentations were important as they allowed the group to share their initial interests, doubts, or queries in relation to the central themes of the summer school. Furthermore, it provided us all with a common vocabulary with which to identify the synergies and tensions between the theoretical framework(s) and the practical activities that took place over the following days.

The first day ended with a visit to the Play Lab at the National Museum of Denmark. This innovative laboratory at the heart of the museum is home to a small team that work on reimagining the relationship between the historical assets and the public in an inherently playful way. This was highly relevant to my own research, and I found their presentation and the subsequent discussion to be highly rewarding, as the Play Lab team were engaged in academic debates around playful experiences in cultural institutions as well as the practicalities of running events and exhibitions on a day-to-day basis.

Day two of the summer school represented a deeper dive into the themes as we explored how play can be thought of conceptually and applied to a variety of practical activities. The first topic was ‘Atmospheres and Ambiances.’ Two of the organisers had transformed the seminar room that we were using entirely by changing the layout of the furniture and introducing interactive artefacts, with different lighting and ambient music added also. We were invited to explore the space in silence as a group and then discussed how these changes had altered our perceptions of the atmosphere both individually and collectively. After this, we split up into small groups and explored the campus through our senses: one member of the group would provide prompts and the other would respond through a constant stream-of-consciousness.

After playing with spaces, we then played with empirical data by writing down our experiences and then editing them in a collaborative, creative, and chaotic way. For example:

I remember becoming intrigued, inspired, interested in the elemental world, the spaces we inhabit, use, appropriate, play with
Jeg husker smagen af min farmors saftevand
Jeg husker følelsen af cykelshorts
​​I remember discussing new methods, new ways of thinking
I remember being told that I was going to have a sister.

 

This activity vividly brought to life many of the ideas discussed on the first day. In the following discussion, there was a general feeling that this playful approach to poetry had enabled an entirely different engagement with our memories than was otherwise possible. I found day two of the summer school to be sometimes challenging, and often I felt outside of my comfort zone, but it certainly provided me with many ways of thinking about the central themes of my doctoral research.

On the final day, we focused mainly on making and tinkering. This was particularly rewarding as it began as quite a simple creative task but evolved into an engaging activity that questioned how we can think about both space and time in playful ways. We ended by presenting back to the group as a whole and then discussing some of the emergent themes of the summer school in a round circle debate. It was interesting how the different backgrounds of the participants had influenced the discussions and produced original thoughts and ideas that were thoroughly interdisciplinary.

As I cycled back from the campus to central Copenhagen, I quite literally bumped into another participant who was going the same way. We decided to grab a drink and reflect on the summer school as a learning experience. Without a doubt, we enjoyed our time, particularly the fact that the PhD students came from a variety of backgrounds which ensured that our debates remained dynamic and multidisciplinary throughout. During our conversation, however, we noted that sometimes these debates produced interesting ideas that could not be fully explored due to the time pressures of the summer school.

On further reflection, I felt this was somewhat unfair as it seems to me that this is the very purpose of attending a summer school as a PhD student: to be challenged and leave with fresh ideas that one’s research can attempt to resolve. My time in Denmark not only provided me with a new network of like-minded researchers, but also allowed me to reengage with my studies from a new playful perspective.

post by Jimiama Mafeni Mase (2018 cohort)

EFI: A Toolbox for Feature Importance Fusion and Interpretation in Python

Divish Rengasamy (a PhD candidate with the Institute for Aerospace Technology at the University of Nottingham, who recently passed his viva) and Dr. Grazziela Figueredo (my supervisor) published a manuscript to Applied Sciences. Their manuscript proposed a solution to address the lack of agreement among feature importance techniques regarding how they quantify the importance of features to machine learning predictions.  Their solution combined the results from multiple feature importance quantifiers to reduce the variance in estimates and to improve the quality of explanations using crisp information fusion techniques, such as mean and majority vote. A few months later, Grazziela scheduled a meeting to discuss opportunities to improve their solution using a fuzzy logic system (FLS), which is one of the main approaches in my PhD due to its capability to capture and model uncertainties and ambiguity in information.  Divish, Dr. Mercedes Torres Torres (who was my second supervisor at that time, but now works for B-Hive Innovations at Lincoln) and I attended the meeting. In the meeting, Divish went through their proposed solution and together we identified some limitations i.e., significant loss of information as their approach reduced several quantifiers to crisp outputs and difficulty to understand the representation of `importance’ as continuous values (also known as importance coefficients). Next, I described the potential benefits of FLSs, such as using human-understandable linguistic terms to define concepts and using fuzzy sets and rules to handle ambiguous information. They loved the idea and provided synthetic data for experiments.

After obtaining the results from the experiments, we observed that FLSs outperformed the previous solution proposed by Divish in capturing increased variation of feature importance caused by increased data dimensionality, complexity and noise. However, the previous solution still showed remarkable performance in situations with less noise and less variability in the importance of features produced by the multiple machine learning models and feature importance techniques.  It was at this point that we decided to develop an open-source Python toolbox called Ensemble Feature Importance (EFI) consisting of the two solutions i.e., crisp information fusion and FLS.  However, we needed an expert in software modularity and testing to handle the development of the toolbox. Fortunately, Aayush Kumar (a software data scientist), who was doing a master’s in data science at the University of Nottingham, agreed to come on board as the developer of the toolbox. Aayush implemented the toolbox in Python programming language as his master’s dissertation. The toolbox consists of packages to automatically optimise ML algorithms, calculate and visualise the importance of features using various feature importance techniques, aggregate the importance of features from multiple ensemble methods, and create fuzzy logic systems for capturing uncertainties and interpreting importance. The audience for the toolbox is machine learning researchers, end-users of machine learning systems, and decision-makers. After developing and testing the toolbox, we approached Professor David Winkler (a Professor of Biochemistry & Chemistry at La Trobe University, Professor of Pharmacy at the University of Nottingham, and Professor of Medicinal Chemistry at Monash University) to assist in the theoretical motivation of the solutions and as a  second pair of eyes in reviewing the paper.

The paper was written in two weeks and the team was divided into two groups. The first group was made up of Divish, Grazziella and I, who developed the paper structure and decided on the content.  We used `Overleaf’, an open-source online collaborative platform for writing and editing documents, to write the paper. The second group was made up of Mercedes, Benjamin Rothwell (Divish’s supervisor) and David, who reviewed and proofread the paper. Comments were made in ‘Overleaf’ using color-coding and the platform’s chat functionality. After several iterations, we all agreed the paper was ready for submission. A month after submission, we received an email titled ‘Final Decision: Notification of Paper’ with delightful content that our paper had been accepted for presentation at the conference. It was indeed a pleasant message because we were aware of the importance of our toolbox to the machine-learning community. There were a few comments from the reviewers. We were required to address all the comments before uploading a camera-ready version of the paper with a maximum of 15 pages including references. I quickly created a google word document for addressing the comments and shared the link with the other authors. We dedicated a week to addressing the comments, as they were minor. Divish and I did a majority of the revision, as we were more familiar with the methodologies. The most recent version of the toolbox is limited to classification tasks; however, we plan to extend the toolbox to deal with regression tasks. In addition, the toolbox is open-source (i.e., accessible to the public for review, modification and extension)  and we encourage other researchers and ML engineers to contribute to its growth, such as improving the structure of the toolbox, implementing additional state-of-the-art ML algorithms and implementing other feature importance techniques.

Finally, I will like to mention the role of this paper in my PhD. To recap, my PhD aims to develop a reliable, comprehensive and interpretable intelligent driving risk assessment system that considers the simultaneous occurrence of driving behaviours and external conditions. The fuzzy approach implemented in the toolbox for aggregating importance coefficients from multiple machine learning models and feature importance techniques, and to make the importance of features easily understandable has also been explored in my thesis. In my thesis, we explore FLSs for capturing uncertainties in driving data and providing meaningful representations of driving behaviours as linguistic terms using fuzzy sets. In addition, we explore human-understandable fuzzy rules to combine the impact of driving behaviours and external conditions on road safety and incorporate expert knowledge into the system.