NVIDIA hackathon

Last week Craig McNeile and Gaurav Ray attended a virtual two day NVIDIA hackathon. The event was organized Dirac. The goal was to learn how to use the new Graphical Processing Units GPUs, which are installed at Cambridge. There were a number of different teams attending the event. At the end of the two days each team made a presentation about the progress made. Below is a screenshot from Gaurav’s talk at the event.

Final year mathematics and statistics student awarded an NIHR pre-doctoral fellowship

Our final year student Emily Prestige, BSc (Hons) Mathematics and Statistics, has been awarded a prestigious two-year National Institute for Health Research (NIHR) pre-doctoral fellowship.

In the final year, Emily did her individual project module to investigate what effects government policies have  on the transmission of COVID-19. She had also studied Medical Statistics and Data Modelling modules in her final year.

The NIHR pre-doctoral fellowship will support her to carry out an MSc in Medical Statistics at London School of Hygiene and Tropical Medicine, as well as conduct research on COVID-19 in the hopes of extending research onto a PhD.

Emily said: “I am incredibly pleased to have been given such a fantastic opportunity and can confidently say that without the support of the fantastic mathematics department at the university I would not be here today. I would like to say a special thank you to my dissertation supervisor Yinghui Wei for all of her support in the last year and to Jenny Sharp who’s guidance over the last four years has been priceless. Finally, I would like to pass on some advice: never give up!”

Online poster day

In the final year, some students choose to study a project module. This is a 40 credit module. The students research a topic that they are particularly interested in, with help from a supervisor. In the time before the pandemic, there was a poster session, at the end of the project. The students explain their research topic, in front of their posters, to staff members and visitors from local businesses.

This year the poster session was online and run on last Thursday. There is a web site with a subset of the posters.

Interview on BBC radio Devon

On Thursday 8th April Dr. Craig McNeile was interviewed on BBC radio
Devon about the new experimental result from the Fermilab National
Laboratory.
The new experimental results were consistent with an
earlier experimental result, but they didn’t agree well with the
theoretical predictions from the standard model of particle physics.
The experimental result is exciting because it may be evidence for new
physics, which may ultimately explain what fundamental particle makes
up dark matter. Dr McNeile does research into the theoretical
calculations required to compare experimental result against theory.

UKLFT Virtual Centre

The UKLFT Virtual Centre will hold its virtual kick-off meeting on Wednesday 24.03.2021 14h-17h GMT.

UKLFT is the STFC-funded Virtual Centre for UK Lattice Field Theory. It currently comprises 35 academics from 10 different UK institutions. Its mission is to: 

  • organise an Annual Meeting with topical talks with speakers from both within the UK and beyond, and a poster/5 minute seminar session to assist younger researchers to introduce their work
  • organise training, ranging from hackathon-style events focussed on the mastery of new architectures or coding techniques, to workshops for reviews and case studies of new algorithms or analysis techniques such as Machine Learning
  • provide T&S support for early career researchers to visit other UK groups to exchange techniques, master code, and foster collaboration
  • fund short-term visitors from overseas, both to extend existing collaborations and to travel to other UK groups to make fresh contacts

The UKLFT Virtual Centre is currently managed by a Board comprised as follows:

  • Christine Davies (Glasgow)
  • Vincent Drach (Plymouth)
  • Simon Hands (Chair, Swansea)
  • Andreas Jüttner (Southampton)
  • Tony Kennedy (Edinburgh)
  • Kurt Langfeld (Liverpool)
  • Arttu Rajantie (Imperial)
  • Christopher Thomas (Cambridge)

At the meeting we will also outline plans for the Virtual Centre going forward, including details of how to apply for travel costs to support you research once the pandemic restrictions are eased, and an informal discussion on future activities such as training and workshops.

Presentations at the British undergraduate conference

Typically a third year student at our school will work on a small project. This will typically be on research topic that the student is interested in. As while as writing a report for their grades, there are opportunities to present the research at a conference.

This year there are three students in Mathematics at Plymouth, who are making presentations at the British Conference of Undergraduate Research, 12 – 13 April 2021. This year the conference will be virtual. The conference is organized the University of Leeds.

The details about the three talks are below:

Title: What effect does government policy have on the reproductive number for COVID-19?

Student Name: Emily Prestige

Abstract:

The aim of this investigation is to determine the effect of government policies on the transmission of COVID-19 during the period 30/01/2020 – 03/01/2021. This investigation uses both qualitative data in the form of policy legislation and journals, and quantitative data in the form of official government pandemic figures. Key factors of successful government approaches are used to assess the effect of the UK government strategy. The measure used to monitor transmission rates is the instantaneous reproduction number Rt. I used summary statistics, data visualisation, and time series representations, to conduct exploratory data analyses. I also investigated the change in testing capacity and positivity rate over time, to account for factors impacting the number of individuals testing positive for COVID-19. To estimate Rt I used a deterministic Susceptible-Infected-Removed model and a stochastic epidemic model, I then used time series models to predict future behaviour of the positivity rate, transmission rate, and number of new cases. From estimating Rt we see that during lockdowns the rate of transmission is reduced. Predictions show that the positivity rate, transmission rate, and number of new cases, will increase if conditions remain the same, i.e., if further mitigation strategies are not implemented. This investigation finds that when policies were communicated clearly, they had a more significant effect on the reduction of Rt. Furthermore, it also finds several areas which can be improved to enhance the effects of policies on reducing transmission rates. 


Title: Analysis on Influencing Factors of Men’s Singles Scores during Winter Olympic Games Cycle

Student Name: Qing Zhang

Abstract:

With the success of Beijing to host the 2022 winter Olympic Games, the main items of the winter Olympics figure skating – also gradually entered people’s field of vision. Figure skating is a highly technical and artistic integration of the ice sports. In combination with the musical accompaniment, the athletes express the musical mood and their own emotions through various ice dance movements, so the sport requires not only the technical ability of the athletes, but also the artistic performance ability of the athletes. This article selects from the International Skating Union (ISU) winter Olympic Games cycle (2014 and 2018) when the man’s world-class competition achievement data statistical analysis, with a number of factors that affect performance results in figure skating men’s as a research perspective, explores how to improve the comprehensive competitiveness of figure skating men’s player of the main ways. The conclusion is that the top factor scores are all the details of the program content, among which the highest score is the performance score; The lower scores were all the details of the technical elements, with the highest score being the jump score. This indicates that the main factors affecting men’s single skating performance in figure skating are program content, including sliding technique, cohesion, performance, arrangement and musical expression. The number and quality of jumps and spins do not significantly affect a competitor’s final result.

Key words: The Winter Olympics; Influencing factors; Regression Analysis; Analysis of Variance; Factor Analysis


Title: Trending Topics and Epidemic Development of the COVID-19 in China: Sentimental Analysis and Visualization

Student name: Yutong Qin

Abstract

In January 2020, COVID-19 broke out in China. During the Spring Festival, because of the high turnover of people and the infectivity of the virus, the number of COVID-19 confirmed cases is rising rapidly. Led by Weibo, Baidu and Toutiao, with the continuous development of the epidemic, the online public opinion is also changing. To a certain extent, the trending topics represent people’s attitudes and views on the development of epidemic situation. The sentimental analysis of trending topics can comprehend people’s reactions and psychology to the changes of national policies and emergencies, so as to play a guiding role in the countermeasures of similar situations in the future.  This paper selects the epidemic data of the NHC of the PRC from January 20 to April 21 in 2020, and the trending topics data of Weibo, Baidu and Toutiao from December 30 of 2019 to April 21 of 2020, to analyze the trending topics’ emotion with the epidemic change. This paper first groups the epidemic development, then turns the word vector of topics into sentence vector. And then uses TF-IDF algorithm to calculate the weight, artificially labels the results with emotion. Sentimental analysis is realized by using CNN, BosonNLP emotion dictionary, and auxiliary dictionary. Finally uses Python and Echarts to make visualization. According to the results of analysis and epidemic data, the paper can get the relationship between public opinions and the development of the epidemic, so as to better comprehend the psychology of public and find out more appropriate countermeasures. 

AMD GPU Hackathon

The researchers, in the Centre for Mathematical Sciences at the University of Plymouth, who use high performance computing for their research mostly use supercomputers run by an organization called Dirac. Because supercomputing is fast moving field , it is important to keep up with new developments in software and hardware. Dirac also organizes training sessions as well.

Last week Craig McNeile and Gaurav Sinha Ray, attended a two day workshop called: AMD GPU Hackathon.  

AMD is a company that develops computer processors. Graphical Processing Units (GPUs) were orginally developed for graphics, but they have found to speed up scientific and machine learning codes.

At the two day event, Craig and Gaurav, with a collaborator from Glasgow, worked on getting some production codes running on the GPUs made by AMD. There was help from AMD consultants through out the day.

Women in Mathematics Day

We are very pleased to announce that our proposal for hosting the London Mathematical Society’s 2021 Women in Mathematics Day has been successful. We look forward to hosting the event at the University of Plymouth.

The organising committee is Yinghui Wei, Nathan Broomhead, Colin Christopher and Daniel Robertz.

More details will follow.