Xiaoyong (Brian) Yuan (@chbrian77) profile picture

Assistant Professor @clemson ECE | PhD @UFCISE | interested in deep learning, security & privacy | currently working on trustworthy machine learning.

Clemson, SC
Joined November 2011
Photo by Xiaoyong (Brian) Yuan (@chbrian77)
Xiaoyong (Brian) Yuan retweeted
Adversarial examples don’t always have to be imperceptible 🥲
A group of San Franciscans realized that they can disable Waymo and Cruise robotaxis by placing a traffic cone on the vehicle's hood. They're now encouraging others to do it: "Hell no, we do not consent to this."
4
24
2,752
Xiaoyong (Brian) Yuan retweeted
As AI adoption continues to grow, it’s clear we need industry security standards so it is safe for everyone. At @Google we’re excited to introduce our Secure AI Framework (SAIF) designed to raise the security bar and reduce risk, together as an industry blog.google/technology/safet…
241
239
44
1,576
154,894
Xiaoyong (Brian) Yuan retweeted
Brian Yuan, AC and CS, has received the 2022 ORAU Ralph E. Powe Junior Faculty Enhancement Award in Mathematics/Computer Sciences. Congratulations, Brian! blogs.mtu.edu/computing/2022… @michigantech @orau #michigantech #computing #computerscience #appliedcomputing
2
1
4
Xiaoyong (Brian) Yuan retweeted
Brian Yuan (CS) is the PI of a $500K National Science Foundation grant titled, “CNS Core: Small: Privacy-Preserving On-Device Intelligence in the IoT Era.” Congratulations, Brian! blogs.mtu.edu/computing/2022… @michigantech @NSF #michigantech #computing #computerscience
2
5
New Research Topic in @FrontSignalPro - showcasing the work of Early Career Researchers across the entire breadth of #ImageProcessing Led by Shiba Kuanar (@ShibaPrasadUTA), Xiaoyong Yuan (@chbrian77) & Pan He Read more here > fro.ntiers.in/XW8q
3
4
Super excited to share our paper "Membership Inference Attacks and Defenses in Neural Network Pruning" accepted at @USENIXSecurity 2022. Extended version: arxiv.org/abs/2202.03335 Code: github.com/Machine-Learning-…
2
2
We observe that neural network pruning disproportionately affects the pruned model's behavior for members and non-members. The increased privacy risk of pruned models is demonstrated based on our proposed membership inference attack.
1
We also propose a new defense mechanism to protect the pruning process by mitigating the prediction divergence.
Xiaoyong (Brian) Yuan retweeted
Python debugging cheat sheet. @HaroldSinnott @SourabhSKatoch
10
229
5
813
Xiaoyong (Brian) Yuan retweeted
9 key CS topics - and the best book and video for learning each of them: https://teapawt.org/t.co/i6qNiDQPpT (credit @BradfieldCS)
11
357
24
1,021
I am happy to share that our paper "Learning Fast and Slow: Propedeutica for Real-Time Malware Detection" has been published in IEEE Transactions on Neural Networks and Learning Systems (TNNLS). ieeexplore.ieee.org/document…
2
4
7
We introduced Propedeutica, a framework for efficient and effective real-time malware detection, leveraging the best of conventional machine learning and deep learning techniques. Thanks, @gracesrm @abedgregio @dseabraoliveira @andyli_uf
1
2
We are organizing the International Conference on Applied Cryptography in Computer and Communications (EAI AC3 2021). The submission deadline is Dec. 15, 2020. #cryptography #AC3 ac3-conference.eai-conferenc…
1
The submission deadline has been extended to Oct. 12. Please don't hesitate to submit your work to NeurIPS-SpicyFL 2020.
We are organizing a NeurIPS workshop on federated learning. Please don't hesitate to submit your recent work to NeurIPS-SpicyFL 2020. The submission deadline is Oct. 2, 2020. icfl.cc/SpicyFL/2020 @NeurIPSConf #DeepLearning #privacy #ArtificialIntelligence #NeuralNetworks
2
1
Xiaoyong (Brian) Yuan retweeted
Stay at 127.0.0.1 Wear a 255.0.0.0
69
1,730
155
6,493
We are organizing a NeurIPS workshop on federated learning. Please don't hesitate to submit your recent work to NeurIPS-SpicyFL 2020. The submission deadline is Oct. 2, 2020. icfl.cc/SpicyFL/2020 @NeurIPSConf #DeepLearning #privacy #ArtificialIntelligence #NeuralNetworks
This Post is from a suspended account. Learn more
1
1
I'm thrilled to announce that I passed my virtual PhD defense yesterday! I am officially a Doctor now! HUGE thank you to my amazing advisors: Dr. Oliver Wu and @andyli_uf, and awesome committee members @kevinrbbutler, @dseabraoliveira, Dr. Jose Principe!!! #PhDone
5
1
10
Xiaoyong (Brian) Yuan retweeted
[New Paper] What happens if you disturb malware execution? Can we use it to "buy time" for ML detectors? We answer this and more in our new IEEE TDSC paper. With @gracesrm @chbrian77 @abedgregio and @dseabraoliveira . Good to work you all! Read it here: ieeexplore.ieee.org/document…