Cohort 4 researcher, Matt Smith, tells us about attending his first research conference;

Matt OsEach year the Ordnance Survey (OS) invites the PhD students it sponsors to its HQ in Southampton for a workshop involving presentations of each student’s current status and work, panel discussion on topics and general networking. There were 22 PhD students present at the workshop, all from different backgrounds and Universities spanning the UK and even including one person who travelled from New Zealand.

Each student had the opportunity to prepare a poster and presentation. My poster focused on the opportunities of my research question and how I was planning to tackle the area, as I don’t have any data yet. My 10-minute presentation echoed this in more detail, giving examples of some of the design methods I want to use and what I have read so far. There was an hour-long poster session on both days which allowed the staff members to come and ask us about our projects. Feedback on my poster was positive. In particular, they found it visually appealing and engaging, however, it perplexed a few as it isn’t what you would expect for a typical PhD poster!

I felt quite unique as I am the only researcher with a ‘pure’ design background and the only one focussing on wearables and actual physical prototyping. From the conversations I had the majority of other researchers were focused on data, ontologies and especially deep/machine learning and forms of automation. As well as meeting other researchers sponsored by OS I was able to speak with a number of staff members, one person in particular, Tim Newman, expressed an interest in the project as he could see possible applications within the walkers’ community.

Picture1

Written by Luke Wilkinson, Cohort 4.

Between 9am of Tuesday 6th March and 9am Wednesday 7th March, the Royal Society of Chemistry (RSC) ran its annual research poster Twitter competition. The idea behind the competition is for researchers to promote their own research area to not only the research community, but to a global audience through Twitter. Entrants had to post their academic research poster on Twitter, and use one of ten hashtags which best describes their area of research within chemistry along with #RSCPoster. Twitter users could then tweet entrants questions about their research. The RSC scientific committee kept an eye on posters which stimulated wide interest and featured innovative research; however prizes were given based on content & accessibility, design and interaction with other Twitter users.

I was one of six entrants from Loughborough University with my research poster; “Is My Circuit Board Wet?”. The competition was a global event, drawing interest from the research community in multiple countries. In the afternoon of the competition, a Twitter account @ErrantScience, well known for making light hearted cartoon abstracts of RSC Poster competition entrants reproduced my poster, putting a comedic spin on the content. This was one of only six posters in the whole competition to receive the Errant Science treatment, which significantly boosted interest in my research. On Friday 9th March, it was announced I had won 1st prize in the materials category, winning £100 prize money and a certificate.

The competition reached a global audience of over 1.9 million people, so to win was a massive achievement. As yet, I am undecided how I will spend the winnings.

Even after the competition had closed, all of Loughborough’s entrants were still answering questions and receiving support from the scientific community which shows that STEM research at Loughborough is thriving.

Below is an iamge of the winning poster and @errantscience treatment. Congratulations Luke!

Rscluke

Cohort 2 researcher Jamal Umer explains his recent trip to University of Manchester....

I attended the workshop on 07th March-2018 at the Chemistry building of University of Manchester. The workshop was organized by the Bruker UK Ltd. During the workshop several topics were quite informative and enabled me to gain insight into valuable knowledge regarding nanoindentation, In-Situ tribological testing in SEM and AFM as useful tool for 3D printing for certain applications.

The talks on the atomic force microscopy enabled me to expand my knowledge on this topic. One new aspect of AFM that was demonstrated at the conference has potential to be incorporated in my later research. Conversations with several speakers directed me to further literature which would make good introduction to further aspect of my research.  

Wen Gu & Shaun Smith, of Cohort 3 & 4 respectively, recently attended the 2-day Future Powertrain Conference (FPC2018). The FPC2018 invites over 500 delegates from both academia and industry to present innovative research ideas and discuss the future of the industry.

This was the first conference for the students since joining the CDT-EI and the main reasons for attending was to first potentially discover new research topics and secondly anchor their own research by considering where they could add their knowledge to the problems in the industry. In addition, the FPC2018 offered networking sessions throughout both days allowing the valuable opportunity for the students to fully immerse themselves in the sector by speaking to experts in the industry and open the door for possible collaboration.

The students highlighted several presentations that were very useful for a variety of reasons. These included, but were not limited to, talks by Prof Colin Garner of Loughborough University who challenged us to consider the most efficient sources of energy and to revisit the debate of “Battery vs Internal Combustion Engine”, Richard Burke of the University of Bath who presented a catalogue of research opportunities and Doug Cross of Leadfoot who provided a talk on the future of autonomous vehicle and if they can keep up with the expectations of the consumer. Furthermore, talks by Dr Yan Wang of Ford and Dr Mark Peckham of Cambustion provided a greater insight of the challenges that face the sector from an industry perspective

Wen Gu presented a poster at the conference to demonstrate his current research. By implementing the machine learning algorithm, he developed a model learning approach to reflect the engine behavior faster and more accurate. The adaptive model predictive control method is employed to dynamically optimise the engine fuel economy based on the developed time-varying model. This approach is promising in improving fuel economy of powertrain systems and leading the way to next generation low carbon engines. At the end of the conference, he won the third prize of poster competition section.

The future powertrain conference returns in 2019 at the National Motorcycle Museum, Solihull, where students will again be attending representing the CDT-EI.

For further information visit futurepowertrains.co.uk/

 

1.0.0.17

Deep Learning Workshop

22 February 2018

To help apply and develop their literature research on Deep Learning, Jake Rankin attended the Deep Learning Workshop. This would both provide experience with applying deep learning and an opportunity to discuss specific considerations with leading practitioners for implementing deep learning systems. The Centre for Modelling and Simulation (CFMS) Hosted a Deep Learning Workshop at the Bristol and Bath Science park, consisting of talks from Cray, Nvidia Deep Learning Institute, TSystemes and CFMS. This was then supported by a workshop that used Nvidia’s Deep Learning GPU Training System (DIGITS). The purpose of this event was to help organisations understand the challenges of implementing deep learning and the presentations were focused on business applications of deep learning and clarifying the mechanisms of deep learning.

Chris Hegarty (Cray, AI Enterprise Sector Development Lead) gave a useful presentation on the hardware and software required for the implementation of Artificial Intelligence Systems, highlighting several useful toolkits such as CNTK, TensorFlow and Caffe2 (All of which are available for use), but also highlighted a word of caution with regards to the data-processing required (Exceeding well above 25 Gflops) and the costs for algorithm training (~$25k). Adam Grzywaczewski (Nvidia, Deep Learning Solution Architect), who also led the Deep Learning Institute workshops, explained the workings behind neural networks very concisely, with focus on Convolutional Neural Networks (CNN), which would be the focus of our workshop.

The Deep Learning Institute workshop applied deep learning techniques to image processing. This was broken down into three lessons:

-          Image Classification with DIGITS

-          Object Detection with DIGITS

-          Neural Network Deployment with DIGITS and TensorFlow

The conference was a useful experience that has helped to focus literature research and consider practical constraints for the project. The networking was a useful insight of business opinions on various deep learning methods and helpful feedback was shared.

For those interested in finding out more about Nvidia DLI and future workshops, please visit www.nvidia.com/dli

We are part of the Digital Economy CDT Network, that supports and develops training activities for postgraduate research students. In early February we were delighted to attend the annual meeting, showcase and careers symposium held at the Digital Catapult, London.

Day 1 was characterised by fruitful and engaging discussions about the range of training activities and support DEN offers and could offer it student members. Attendees eagerly awaited news about the Summer School 2018, being hosted this year by the Centre for Digital Entertainment and with topics such as AI and machine learning on the agenda it was clear this year’s event was going to draw a big crowd. Day 1 concluded with a CDT showcase, allowing the member CDT’s to demonstrate the quality and quantity of research taking place at their centres and for the researchers to find out more about their peers across the UK. A particular highlight was the VR game simulation from the IGGI CDT team!

Careers was the focus of the following day and saw guest speakers, all who were alumni from CDT members of DEN. Each speaker shared their career journey, the decisions, complications and successes achieved along the way. There were speakers from academia and industry with both SME’s and large enterprises represented. The talks gave a realistic insight to what lies on the other side of the PhD thesis and each speaker shared their thoughts on what a doing a PhD in a CDT has meant to them and what they would do differently (finishing the writing process before getting a job was a reoccurring theme!).

Den Showcase

The West University of Timisoara hosted an international winter school on one of the most dynamic and important sectors of computer science, Big Data. The event covered a broad spectrum of industrial research and innovation with potential over scientific discoveries, medicine, engineering, business models and society. Attendees were master students, PhD students and industry practitioners from over 30 different countries (most participants from outside the country being from USA, Italy and German).  Among the tutors leading the specialised sessions were Jeffrey Ullman, Stanford University, and Bing Liu, University of Illinois, Chicago.

Cohort 3 research, Orange Gao, attended the BigDat winter school to update herself about the most recent advances in the area of big data and to know current research and industrial innovation in different fields. Her major is computer science, specifically on computer vision, and there are many common algorithms with big data, so she hoped the course would help her to broaden her knowledge and understand real applications in commercial world to prepare the future career choices. She chose 3 courses, which was running in parallel during the whole event. Semantic Technologies for Open Data Publishing, which was a fantastic introduction to ontology-based data management (OBDM), Network Science Fundamentals, in which she deep-dived into recent advances in stochastic flow clustering of networks and Lifelong Learning and its Applications in NLP, which is an advanced machine learning paradigm. Since attending the winter school, her future plans are to, firstly, sort out the notes which she kept during the courses! The biggest gain she sees from BigDat was the change from thinking towards to the knowledge. She explains that researchers should not only digest necessary domain knowledge but also understand knowledge’s dynamics and constantly provide technical vision.

Several CDT-EI researchers from Heriot-Watt University are undertaking a 6-month training programme in applied data science. The course, which is being run by Cambridge Spark Ltd will take place over 20 weekends from February 2018 to July 2018. The programme will be project-based learning over 9 modules covering topic areas such as cloud computing and databases, neural networks and deep learning. The researchers will also undertake an industry sponsored project as well as attend a final showcase weekend. Lucas Kirschbaum, CDT-EI, Cohort 4 says

The data science course I will attend is an elaborate training in the most recent data processing techniques. Therefore, this course will not only build up fundamental knowledge but also support me understanding the practical aspects of managing the data I collect within the scope of my PhD. I am hoping to gain valuable insights of how to set up customized algorithms for different testing scenarios, validating the data and in general how to approach big datasets.

follow us
@cdt_ei

2 hours ago RT @WeKnowRobots: 📣Call for Papers and Registration 📣 2nd Robotics and Autonomous Systems Conference for PhD students & early-career resear…
2 weeks ago Supporting the Charnwood Orchestra and second violin Mary from @WolfsonSchool tonight for their winter concert… https://t.co/Zj9Y8AY5H9
1 month And @CDT_EI is one of them too! https://t.co/GJkTzCBlr0

General Enquiries Contacts:

Loughborough University

Loughborough, Leicestershire
LE11 3TU
United Kingdom
Tel: +44 (0)1509 263171
Dr Donna Palmer
cdt-ei@lboro.ac.uk

Heriot-Watt University

Edinburgh, Scotland
EH14 4AS
United Kingdom
Tel: +44 (0)131 449 5111
Dr Keith Brown
k.e.brown@hw.ac.uk