Adversarial Learning in RecSys

Page dedicated to Research on Adversarial Machine Learning for Recommendation and Search. #TrustworhyML #Attack #Defense #Privacy. Tag the page to be retweeted!

[ELSA - European Lighthouse on Secure and Safe AI @elsa_lighthouse]

Today, there was a fascinating workshop on Technical Robustness& Safety on AI led by @biggiobattista. Battista is Associate Professor at the @univca and co-founder of the cybersecurity company @pluribus_one . We are grateful for our various partners in the ELSA project.

[Balázs Hidasi @balazshidasi]

What a first day of #recsys2023 proper! Talked with several people & presented one of our works. Glad many reacted positively. My highlight from the day is SharpCF presented by @vivwylai from Visa. Good idea, clean execution & may also improve other algos. https://t.co/TWN6Ivpca5

[Daniele Malitesta @dmalitesta]

Our preprint "Formalizing #Multimedia #Recommendation through #Multimodal Deep Learning" is now out on arXiv!

https://t.co/p548XU81kw

(1/2)

[Felice Antonio Merra @merrafelice]

Read our last paper @SIGIRConf "Denoise to protect: a method to robustify visual recommenders from adversaries" to get insights on how to defend a #RecSys #IR engine from malicious product images presented today!
🔗https://t.co/QK81vncBHl
🧑‍💻 @sisinflab

[Alberto Carlo Maria Mancino @alberto_mancino]

-11 days to #SIGIR2023 🚀
I will present a novel defense strategy for robustify visual recommenders.

If you are interested feel free to take a look at our work or get in touch with us @merrafelice @dmalitesta @walteranelli @TommasoDiNoia

https://t.co/o5qcMy4L6M

[Felice Antonio Merra @merrafelice]

Clean product images to reduce the risk of malicious users!

Check our work in #Adversarial #RecSys that shortly will be presented at #SIGIR2023!

[Felice Antonio Merra @merrafelice]

(1/3) Too excited to share that the last work on my Ph.D. dissertation on #Adversarial #RecSys "Denoise to protect: a method to robustify visual recommenders from adversaries" has been accepted as Short Paper at @SIGIRConf #SIGIR2023.

📎 Pre-print soon!

[Trustworthy ML Initiative (TrustML) @trustworthy_ml]

📢 Are you on the job market this year or looking for new opportunities, whether full-time, internships, postdocs, etc? We would love to promote your work.

Write a tweet describing your work and tag us (@trustworthy_ml), and we will retweet you! :)

[Machine Learning Security Laboratory @mlsec_lab]

We are excited to present a new event in our seminar series on ML Security! We will host @EdwardRaffML (@BoozAllen ) on February 16, 2023, at 16:00 CET.

Registration: https://t.co/cyu00M8HMV

@adversarial_ML @trustworthy_ml @aivillage_dc @RedTeamVillage_

[Eric Wallace @Eric_Wallace_]

Models such as Stable Diffusion are trained on copyrighted, trademarked, private, and sensitive images.

Yet, our new paper shows that diffusion models memorize images from their training data and emit them at generation time.

Paper: https://t.co/LQuTtAskJ9

👇[1/9]

[Markus Schedl @m_schedl]

If you are looking for new challenges in 2023, consider applying for a university assistant/postdoc position (40h/week, 6 years, on topics of RecSys, IR, NLP, MM, Fairness, etc.) in our group at @jkulinz @cpjku https://t.co/eimbfwIjO8 @ACMRecSys @SIGIRConf @SIGIR2013 Please RT

[Pin-Yu Chen @pinyuchenTW]

While it's true that you can watch our @NeurIPSConf tutorial recording at any time, our Q&A (in 50 mins) and panel (in 60 mins) will be live! Come join us to discuss foundation models
@RisingSayak @sijialiu17 @payel791 @AlexGittens8 @RTFMCelia @uiuc_aisecure @HildeKuehne

[Yashar Deldjoo @yashardel]

Special Issue on "Trustworthy Recommender System" has a deadline in January 15, 2023. Strong works on #security #privacy, #explainability and #fairness of recommender systems, and conversational agents are welcome to submit their work to #ACM_TORS.

[Srijan Kumar - Lighthouz AI (YC S24) 🤖 @srijankedia]

My Ph.D. student Sejoon Oh @GTCSE presenting his paper on stability of recommender systems at ACM @cikm2022 #CIKM2022

Paper link: https://t.co/opd7YcIAtL

[Trustworthy ML Initiative (TrustML) @trustworthy_ml]

📢 1/ Call for Proposals: Research Brainstorm on “The Future of Trustworthy ML” at the Trustworthy ML Initiative Symposium on 10/27.

Details 👇

[Pin-Yu Chen @pinyuchenTW]

Very exciting and proud work on scaling up robust (adversarial) training via efficient distributed optimization from an amazing team and collaborators! Special kudos to @sijialiu17 for his leadership to make all these great things happen @IBMResearch @MITIBMLab @UncertaintyInAI

[Adversarial Learning in RecSys @AdversRecSys]

"TUTORIAL ON ADVERSARIAL ROBUSTNESS OF DEEP LEARNING" live in less than 2 hours at @CIKM2021 by Dr. Wenjie Ruan, @XinpingYi, and Dr. Xiaowei Huang.

Tutorial WebSite: https://t.co/X6Ia3h3yeU

#Adversarial #Robustness #CIKM2021 #DeepLearning

[Adversarial Learning in RecSys @AdversRecSys]

2 papers in the main track at #RecSys2022 @ACMRecSys

Adversary or Friend? An adversarial Approach to Improving Recommender Systems
by Shivaswamy and Dario Garcia (@NetflixResearch)

Defending Substitution-based Profile Pollution Attacks on Sequential Recommenders by Yue et al.

[Adversarial Learning in RecSys @AdversRecSys]

Paper recommendation: https://t.co/QzbwJzXheh @kdd_news

[Zihan Wang @ZihanWa54274484]

Our full paper "Debiasing Learning for Membership Inference Attacks Against Recommender Systems" is now online. #KDD2022

Co-authors: @jayren3 @zhaochun_ren @mdr @fei__sun

[Ambra @ambrademontis]

The paper submission deadline for the 15th ACM workshop on Artificial Intelligence and Security (https://t.co/H4qP2Ml6pH), co-located with @acm_ccs, is postponed to July 20!

#MLsec #MachineLearning #AI #adversarial @adversarial_ML #malware @trustworthy_ml @RedTeamVillage_