Imperial NetSys
@Imperial_NetSys
Followers
84
Following
97
Media
12
Statuses
81
Networks and Systems Lab at the Department of Computing and I-X, Imperial College London @ICComputing @ImperialX_AI @Imperial_ISST @ImperialCollege
Kensington
Joined October 2022
This week @TheOfficialACM @ACMMobiCom 2025, Josh Millar will present his paper "Benchmarking Ultra-Low-Power μNPUs" , joint work another with folks at Imperial and Cambridge: https://t.co/mM5xaek0rY
arxiv.org
Efficient on-device neural network (NN) inference offers predictable latency, improved privacy and reliability, and lower operating costs for vendors than cloud-based inference. This has sparked...
0
0
0
Appearing at @PET_Symposium next week, work from our group at @ImperialX_AI : “TeleSparse: Practical Privacy-Preserving Verification of Deep Neural Networks” Paper: https://t.co/w7yUopcqwf Code:
github.com
verifiable and privacy-preserving inference via sparse neural networks by ZK-SNARKs schema (halo2) - mammadmaheri7/TeleSparseRepo
0
1
2
Following Prof. Hamed Haddadi’s (@ICComputing & I-X) fantastic Inaugural Lecture last week, we are excited to share our latest Research Spotlight delving into his research on IoT, Privacy & Security, and Human-Centred Systems. @realhamed @Imperial_NetSys
https://t.co/bzalcv0MId
0
5
16
Event News! Prof. Hamed Haddadi (@ICComputing & I-X) is organising the Multi-Service Networks workshop, focusing on the latest trends in networks and systems. 🌐 🗓 8 January 🕐 1-5 pm 📍 Huxley Building (@imperialcollege) https://t.co/M57j1ivVG9
@realhamed @Imperial_NetSys
eventbrite.co.uk
Welcome to the winter edition of Multi-Service Networks workshop, where we will cover all aspects of networked systems.
0
1
7
Congratulations to @YushanStar for having won the Best Paper Award at @AcmSensys HumanSys2024. work from his internship at @BellLabs
0
0
5
📢 Our new paper "Low-Energy On-Device Personalization for MCUs," will be presented at ACM/IEEE SEC'24 (@ACM_IEEE_SEC) I worked on this research with @RanyaAloufi, Xavier Cadet, @yc_york, @PBarnaghi, @realhamed 📰Paper: https://t.co/3eRgMfm3CO 💻Code : https://t.co/MGPJIFSWgw
0
3
5
Apply for the I-X WAI Early Career Grant! 🌟 The grant supports women and non-binary individuals engaged in #AI or wanting to enter the field. If you are a Master’s or PhD student or a Postdoc at @imperialcollege, apply now! 🤖 🗓️ Deadline: 31 August 🔗 https://t.co/UtLb2JnLQE
1
7
11
Great presentation by @j0shmillar about his ongoing work on energy-efficient ML at #Coseners24
1
0
4
@J_P_M_Morgan @e_giunchiglia @Ed__Johns @realhamed Prof. Hamed Haddadi (@realhamed) is showcasing the Internet of Things (#IoT) lab. 🌐
1
1
5
Some of our research being presented today at the #IXOpenDay
Welcome to our live stream of #IXOpenDay. Throughout the day, we’ll be reporting on the event. If you’re attending, please tag us and use our hashtag. We’d love to hear from you! 🎉
0
0
1
We are delighted to announce that @imperialcollege has been ranked 2nd in the world and 1st in the UK and Europe in this year’s QS @worlduniranking Congratulations to #OurImperial community on this brilliant result! #Computing #Stem #university
1
5
37
“Mitigating Hallucinations in Large Language Models via Self-Refinement-Enhanced Knowledge Retrieval”, new work from @NiuMengjia to appear at GenAI workshop at @SIGIRConf
0
0
3
Looking forward to attending the Confidential Computing Summit @confcompsummit this week! Presenting a poster about our ongoing research in trusted and private on-device learning. More details: https://t.co/px7CqCnZfs If you're around and want to know more, come say hi!
comet-cc.github.io
Confidential and Open Machine Learning with Enhanced Trust
#CCSummit is just 3 days away! Accenture, Microsoft Azure, Google, Anthropic, and more…From C-Suite execs to engineers, all the big players in #confidentialcomputing will share what’s next #datasecurity #generativeAI. Register now - It’s not too late! https://t.co/zY2017d4dJ
1
3
5
𝘐 𝘢𝘮 𝘵𝘳𝘶𝘭𝘺 𝘩𝘰𝘯𝘰𝘶𝘳𝘦𝘥 𝘵𝘰 𝘫𝘰𝘪𝘯 𝘵𝘩𝘦 𝘌𝘹𝘦𝘤𝘶𝘵𝘪𝘷𝘦 𝘚𝘵𝘦𝘦𝘳𝘪𝘯𝘨 𝘉𝘰𝘢𝘳𝘥 𝘰𝘧 𝘵𝘩𝘦 𝘐𝘰𝘛𝘚𝘍 😎👩💻🦋 @ucleeenews @UCL_ICCS @imperial_isst @PETRASiot
https://t.co/VVT1KC6kX0
3
3
35
Paper: https://t.co/77YyrbEtV3… Authors: @AliShahinShams1 @GefeiTan @TCebere @aurelien_bellet* @realhamed* @NicolasPapernot* @wangxiao1254* @adrian_weller* #ICLR2024
@iclr_conf
0
1
6
How do you know if an ML model was trained privately? Confidential-DPproof, our spotlight #ICLR2024 paper, allows institutions to directly prove to you through the execution of a Zero Knowledge Proof protocol that they train ML models in a privacy-preserving manner. 1/n
1
12
30
I'm looking for a postdoc to join my group at @NYUAbuDhabi in sunny Abu Dhabi this fall! ☀️🏖️💻Apply here: https://t.co/GbJqNJfPeo Three good reasons to apply: 1/2
2
11
27
Two papers from NetSys / @ImperialX_AI at @euromlsys today: @sansib presenting "Towards Attestable and private ML with CCA” https://t.co/Q6xXR3XwIB
@j0shmillar presenting "Towards Low-Energy Adaptive Personalization for Resource-Constrained Devices"
arxiv.org
The personalization of machine learning (ML) models to address data drift is a significant challenge in the context of Internet of Things (IoT) applications. Presently, most approaches focus on...
1
3
8
Chat with us at @euromlsys later this month, and stay tuned for (longer) follow-up work in this direction! 4/4
0
1
2
As CCA is still under development, to facilitate further research in this space, we're releasing code and a setup guide. Check out our GitHub link above! 3/4
1
1
3