Centre for Doctoral Training in Cloud Computing for Big Data, Newcastle University
Newcastle Upon Tyne, England
Joined June 2014
- Tweets1K
- Following423
- Followers1.2K
- Likes956
CDT in Cloud Computing for Big Data retweeted
Fantastic time at IEEE Conference on Games in Boston last week, presenting our DRL Gym wrapped Unity simulation SWiMM DEEPeR. @ieee_cog thank you for the brilliant organisation, content and generous catering! @ccbdcdt thank you for sending us ! Look forward to sharing the paper
CDT in Cloud Computing for Big Data retweeted
It's that time again!
Join us for the 7th UK Systems Research Challenges workshop!
Be part of our friendly & inclusive community to discuss topics affecting the design & implementation of large-scale systems
Abstract deadline: 27 March 2023
Submission: t.ly/tgsP
Only a few days left to apply!
@GeospatialCDT - one for your students?
Are you a #PhD or #ECR looking to apply #DataScience skills to real problems?
Applications for our @UniofNewcastle x @LIDA_UK x @turinginst Data Study Group close on Friday!
Amazing challenges from @LIDA_UK @AkzoNobelUK @theriverstrust & @Northpowergrid
More info 👇
Exciting opportunity for #PGR students across the @turinginst network!
We're hosting a @turinginst Data Study Group at @TheCatalystUK, in collaboration with @LIDA_UK
Develop your skills while working alongside talented researchers from multiple disciplines & help to solve a real-world #datascience challenge
More info👉t.ly/5JHP
CDT in Cloud Computing for Big Data retweeted
This afternoon we are launching our new website. Please visit geospatialcdt.ac.uk for information on our programme, studentships and the amazing work our students are doing.
More to follow as we report on the continued success of our researchers.
In our latest preprint we use Deep Learning to predict the status of subjects (patient with mitochondrial disease or healthy control) from Image Mass Cytometry image stacks captured from human muscle tissue biopsies.
arxiv.org/abs/2210.17360
A different kind of morning, filming at Blyth Marine Station for an upcoming series profiling some of the amazing work taking place across @Sage_NCL
Also ft Freddy Whippet, who tried very hard to be 'helpful' all morning
CDT in Cloud Computing for Big Data retweeted
I was lucky enough to win this very cool gadget at #RHResearchDay today in Brno, Czech Republic! 😍 It's a locally-designed ESP32-based educational board with lots of NeoPixels, beeper and controls. I think it'd make a pretty cool Tenori-on. Děkuji, @redhatcz! 🇨🇿
We're here at the #RHResearchDay in Brno! Great to hear about the amazing work that's going on between Red Hat and academia
CDT in Cloud Computing for Big Data retweeted
At #SMM2022 this week? Want to learn about speeding up photo-id using computer vision? I'll be presenting our work on that very topic today 6-7:30pm in the poster hall! Can't make it? I'll be back again on thursday 10:30am-12pm, and available to chat here or the conf site!
ALT A copy of my SMM22 poster, entitled Automatic Photo-Identification of Cetaceans Using Deep Learning. OVERVIEW: Multi stage pipeline of trained neural networks Automatically detect fins via a Mask R-CNN network using pixel overlap Detected fins are cleaned then sent for catalogue matching via a Siamese Neural Network (SNN) SNN creates a numeric representation of the fin (Fin Embeddings) Comparison of representation allows for catalogue matching Can be used to flag possible previously unseen individuals RESULTS: System components tested on catalogues covering a range of locations, times, and species to ensure generalisability. Detector (Mask-RCNN): Tursiops aduncus, Tanzania: 91% pixel overlap. Total # fins = 616 T. truncatus & Lagenorhynchus albirostris, UK: 96% pixel overlap. # fins = 2900 Catalogue Matcher (SNN): T. truncatus & L. albirostris, UK: 83% ID within 10 MLM. # of classes = 44 T. truncatus, USA: 97% ID within 10 MLM. # of classes = 23
Congratulations also to @ppintodsilva (cohort 3), who successfully defended his thesis 'Knowledge Discovery in Vehicle Identification Sensor Networks'!
Pedro was supervised by @MattForshaw & @stevemcgough, and examined by Nigel Thomas and @SpaceTimeJames
We're excited that @ppintodsilva will be returning to @UniofNewcastle in September in a teaching capacity
Congratulations to @RobGeada (cohort 4) who successfully defended his thesis, 'Neural Architecture Search Across Expanded and Infinite Spaces: Efficient Design of Optimal Neural Networks'!
Rob was supervised by @stevemcgough, and examined by @jaumebp and Elliott Crowley
CDT in Cloud Computing for Big Data retweeted
Episode 2 on "Efficient Answering of Historical What-if Queries" with the excellent Felix Campbell is coming tomorrow at 9AM BST! @IITDBGroup @illinoistech @LordPretzel
#DataManagement #DataProvenance #WhatIfQueries #SIGMOD
We're loving this podcast from @jwaudberry - fantastic initiative to bridge the gap between research and industry
Each series will focus on a different conference, the first is on papers published at ACM @sigmod /PODS '22
🚨 New episode 🚨
Alex Isenko | Where Is My Training Bottleneck? Hidden Trade-Offs in Deep Learning Preprocessing Pipelines | #1 - disseminatepodcast.podcastpa…
via @podcastpage
We definitely picked an excellent day to be by our beautiful coastline 😎
Fantastic input from everyone and great to take stock of where we are and where we're heading
Very productive @ccbdcdt Centre for Doctoral Training All Hands at @UniofNewcastle 's Dove Marine Laboratory - a nice day to be beside the seaside.
Already looking forward to the next one - big shout out to @OpenLab_Ncl for organising this one and bringing us along!
Writing retreats are amazing for improving focus and we're always amazed at how much we achieve in a few days
🖊️We had another writing retreat and this time we joined up with our upstairs neighbors the @ccbdcdt.
openlab.ncl.ac.uk/posts/2022…
