{"type":"ScholarlyArticle","title":"Tolerating Outliers: Gradient-Based Penalties for Byzantine Robustness and Inclusion","authors":[{"name":"Latifa Errami","slug":"latifa-errami","given":"Latifa","family":"Errami","affiliation":"College of Computing , Mohammed VI Polytechnic University , Ben Guerir , Morocco"},{"name":"El Houcine Bergou","slug":"el-houcine-bergou","given":"El Houcine","family":"Bergou","affiliation":"College of Computing , Mohammed VI Polytechnic University , Ben Guerir , Morocco"}],"abstract":"This work investigates the interplay between Robustness and Inclusion in the context of poisoning attacks targeting the convergence of Stochastic Gradient Descent (SGD). While robustness has received significant attention, the standard Byzantine defenses rely on the Independent and Identically Distributed (IID) assumption causing their performance to deteriorate on non-IID data distributions, even without any attack. This is largely due to these defenses being excessively cautious and discarding benign outliers. We introduce a penalty-based aggregation that accounts for the discrepancy between trusted clients and outliers. We propose the use of Linear Scalarization (LS) as an enhancing method to enable current defenses to simultaneously circumvent Byzantine attacks while also granting inclusion of outliers. This empowers existing defenses to not only counteract malicious adversaries effectively but also to incorporate outliers into the learning process. We conduct a theoretical analysis to demonstrate the convergence of our approach. Specifically, we establish the robustness and resilience of our method under standard assumptions. Empirical analysis further validates the viability of the proposed approach. Across mild to strong non-IID data splits, our method consistently either matches or surpasses the performance of current approaches in the literature, under state-of-the-art Byzantine attack scenarios.","doi":"10.24963/ijcai.2024/435","url":"https://ijcai.org/proceedings/2024/435","pdf":"https://ijcai.org/proceedings/2024/0435.pdf","markdown":"https://ijcai.org/proceedings/2024/435.md","bibtex":"https://ijcai.org/proceedings/2024/bibtex/435","volume":{"slug":"2024","title":"Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2024","url":"https://ijcai.org/proceedings/2024/","year":2024,"location":"Jeju, Korea","dates":"3-9 August 2024","editors":"Kate Larson","isbn":"978-1-956792-04-1","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2024"},"section":"Machine Learning","track":"Main Track","number":435,"pages":{"first":3935,"last":3943},"keywords":["Machine Learning: ML: Robustness","AI Ethics, Trust, Fairness: ETF: Fairness and diversity","Machine Learning: ML: Trustworthy machine learning"],"videos":[],"references":[{"position":1,"raw":"[Alistarh et al., 2018] Dan Alistarh, Zeyuan Allen-Zhu, and Jerry Li. Byzantine stochastic gradient descent. In S. Ben- gio, H. Wallach, H. Larochelle, K. Grauman, N. Cesa- Bianchi, and R. Garnett, editors, Advances in Neural In- formation Processing Systems, volume 31.","title":"Byzantine stochastic gradient descent","authors":"Dan Alistarh, Zeyuan Allen-Zhu, Jerry Li","year":null,"venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Allouah et al., 2023] Youssef Allouah, Sadegh Farhad- khani, Rachid Guerraoui, Nirupam Gupta, Rafael Pinot, and John Stephan. Fixing by mixing: A recipe for opti- mal byzantine ml under heterogeneity. In Francisco Ruiz, Jennifer Dy, and Jan-Willem van de Meent, editors, Pro- ceedings of The 26th International Conference on Artifi- cial Intelligence and Statistics, volume 206 of Proceedings of Machine Learning Research, pages 1232-1300. PMLR, 25-27 Apr 2023.","title":"Fixing by mixing: A recipe for optimal byzantine ml under heterogeneity","authors":"Youssef Allouah, Sadegh Farhadkhani, Rachid Guerraoui, Nirupam Gupta, Rafael Pinot, John Stephan","year":"2023","venue":"Proceedings of The 26th International Conference on Artificial Intelligence and Statistics","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Baruch et al., 2019] Gilad Baruch, Moran Baruch, and Yoav Goldberg. A little is enough: Circumventing de- fenses for distributed learning. In Advances in Neural In- formation Processing Systems, volume 32.","title":"A little is enough: Circumventing defenses for distributed learning","authors":"Gilad Baruch, Moran Baruch, Yoav Goldberg","year":null,"venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Blanchard et al., 2017] Peva Blanchard, El Mahdi El Mhamdi, Rachid Guerraoui, and Julien Stainer. Machine learning with adversaries: Byzantine tolerant gradient de- scent. Advances in Neural Information Processing Sys- tems, 2017-Decem, 2017.","title":"Machine learning with adversaries: Byzantine tolerant gradient descent","authors":"Blanchard","year":"2017","venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Boussetta et al., 2021] Amine Boussetta, El Mahdi El- Mhamdi, Rachid Guerraoui, Alexandre Maurer, and Sébastien Rouault. AKSEL: Fast Byzantine SGD. Leibniz International Proceedings in Informatics, LIPIcs, 184(8), 2021.","title":"AKSEL: Fast Byzantine SGD","authors":"Amine Boussetta, Mahdi El, Rachid El-Mhamdi, Alexandre Guerraoui, Sébastien Maurer, Rouault","year":"2021","venue":"Leibniz International Proceedings in Informatics","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Cao et al., 2021] Xiaoyu Cao, Minghong Fang, Jia Liu, and Neil Zhenqiang Gong. Fltrust: Byzantine-robust federated learning via trust bootstrapping, 2021.","title":"Fltrust: Byzantine-robust federated learning via trust bootstrapping","authors":"Xiaoyu Cao, Minghong Fang, Jia Liu, Neil Zhenqiang, Gong","year":"2021","venue":null,"doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Damaskinos et al., 2018] Georgios Damaskinos, El Mahdi El Mhamdi, Rachid Guerraoui, Rhicheek Patra, and Mahsa Taziki. Asynchronous byzantine machine learning (the case of sgd). 35th International Conference on Machine Learning, ICML 2018, 3, 2018.","title":"Asynchronous byzantine machine learning (the case of sgd)","authors":"Georgios Damaskinos, Mahdi El El, Rachid Mhamdi, Rhicheek Guerraoui, Mahsa Patra, Taziki","year":"2018","venue":"35th International Conference on Machine Learning, ICML 2018","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[El Mhamdi et al., 2018] El Mahdi El Mhamdi, Rachid Guerraoui, and Sebastien Rouault. The hidden vulnera- bility of distributed learning in byzantium. In 35th Inter- national Conference on Machine Learning, ICML 2018, volume 8, 2018.","title":"The hidden vulnerability of distributed learning in byzantium","authors":"El Mhamdi","year":"2018","venue":"35th International Conference on Machine Learning, ICML 2018","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[El-Mhamdi et al., 2021a] El Mahdi El-Mhamdi, Sadegh Farhadkhani, Rachid Guerraoui, Arsany Guirguis, Lê- Nguyên Hoang, and Sébastien Rouault. Collaborative learning in the jungle (decentralized, byzantine, heteroge- neous, asynchronous and nonconvex learning). In M. Ran- zato, A. Beygelzimer, Y. Dauphin, P.S. Liang, and J. Wort- man Vaughan, editors, Advances in Neural Information Processing Systems, volume 34.","title":"Collaborative learning in the jungle (decentralized, byzantine, heterogeneous, asynchronous and nonconvex learning)","authors":"El-Mhamdi","year":"2021","venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"Curran Associates, Inc., 2021.","title":null,"authors":"Curran Associates, Inc","year":"2021","venue":null,"doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[El-Mhamdi et al., 2021b] El-Mahdi El-Mhamdi, Rachid Guerraoui, and Sébastien Rouault. Distributed Momen- tum for Byzantine-resilient Stochastic Gradient Descent. International Conference on Learning Representations, 2021.","title":"El-Mahdi El-Mhamdi","authors":"El-Mhamdi","year":"2021","venue":"Distributed Momentum for Byzantine-resilient Stochastic Gradient Descent. International Conference on Learning Representations","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Farhadkhani et al., 2022] Sadegh Farhadkhani, Rachid Guerraoui, Nirupam Gupta, Rafael Pinot, and John Stephan. Byzantine machine learning made easy by resilient averaging of momentums. In Kamalika Chaud- huri, Stefanie Jegelka, Le Song, Csaba Szepesvari, Gang Niu, and Sivan Sabato, editors, Proceedings of the 39th International Conference on Machine Learning, volume 162 of Proceedings of Machine Learning Research. PMLR, 17-23 Jul 2022.","title":"Byzantine machine learning made easy by resilient averaging of momentums","authors":"Sadegh Farhadkhani, Rachid Guerraoui, Nirupam Gupta, Rafael Pinot, John Stephan","year":"2022","venue":"Proceedings of the 39th International Conference on Machine Learning","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Fu et al., 2019] Shuhao Fu, Chulin Xie, Bo Li, and Qifeng Chen. Attack-resistant federated learning with residual- based reweighting. CoRR, abs/1912.11464, 2019.","title":"Attack-resistant federated learning with residualbased reweighting","authors":"Fu","year":"2019","venue":null,"doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Gorbunov et al., 2023] Eduard Gorbunov, Samuel Horváth, Peter Richtárik, and Gauthier Gidel. Variance reduction is an antidote to byzantines: Better rates, weaker assump- tions and communication compression as a cherry on the top. In International Conference on Learning Representa- tions, 2023.","title":"Variance reduction is an antidote to byzantines: Better rates, weaker assumptions and communication compression as a cherry on the top","authors":"Gorbunov","year":"2023","venue":"International Conference on Learning Representations","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Jayanth Regatti, 2022] Abhishek Gupta Jayanth Regatti, Hao Chen. Byzantine resilience with reputation scores. In Annual Allerton Conference on Communication, Control, and Computing, 2022.","title":"Byzantine resilience with reputation scores","authors":"Abhishek Gupta, Jayanth Regatti, Hao Chen","year":"2022","venue":"Annual Allerton Conference on Communication, Control, and Computing","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Karimireddy et al., 2021] Sai Praneeth Karimireddy, Lie He, and Martin Jaggi. Learning from history for byzantine robust optimization. In Marina Meila and Tong Zhang, editors, Proceedings of the 38th International Conference on Machine Learning, volume 139 of Proceedings of Ma- chine Learning Research. PMLR, 18-24 Jul 2021.","title":"Sai Praneeth Karimireddy, Lie He, and Martin Jaggi. Learning from history for byzantine robust optimization","authors":"Karimireddy","year":"2021","venue":"Proceedings of the 38th International Conference on Machine Learning","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Karimireddy et al., 2022] Sai Praneeth Karimireddy, Lie He, and Martin Jaggi. Byzantine-robust learning on het- erogeneous datasets via bucketing. In International Con- ference on Learning Representations, 2022.","title":"Byzantine-robust learning on heterogeneous datasets via bucketing","authors":"Praneeth Sai, Lie Karimireddy, Martin He, Jaggi","year":"2022","venue":"International Conference on Learning Representations","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Krizhevsky and Hinton, 2009] Alex Krizhevsky and Geof- frey Hinton. Learning multiple layers of features from tiny images. Technical report, University of Toronto, Toronto, Ontario, 2009.","title":"Learning multiple layers of features from tiny images","authors":"Alex Krizhevsky, Geoffrey Hinton","year":"2009","venue":null,"doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Krizhevsky et al., 2017] Alex Krizhevsky, Ilya Sutskever, and Geoffrey E. Hinton. Imagenet classification with deep convolutional neural networks. Commun. ACM, 60(6):84-90, may 2017.","title":"Imagenet classification with deep convolutional neural networks","authors":"Alex Krizhevsky, Ilya Sutskever, Geoffrey E Hinton","year":"2017","venue":"Commun. ACM","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Lamport et al., 1982] Leslie Lamport, Robert Shostak, and Marshall Pease. The byzantine generals problem. ACM Trans. Program. Lang. Syst., 4(3):382-401, jul 1982.","title":"The byzantine generals problem","authors":"Leslie Lamport, Robert Shostak, Marshall Pease","year":"1982","venue":"ACM Trans. Program. Lang. Syst","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[LeCun et al., 1998] Yann LeCun, Léon Bottou, Yoshua Bengio, and Patrick Haffner. Gradient-based learning ap- plied to document recognition. Proceedings of the IEEE, 86(11):2278-2324, 1998.","title":"Gradient-based learning applied to document recognition","authors":"Yann Lecun, Léon Bottou, Yoshua Bengio, Patrick Haffner","year":"1998","venue":"Proceedings of the IEEE","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Li et al., 2019] Liping Li, Wei Xu, Tianyi Chen, Geor- gios B. Giannakis, and Qing Ling. RSA: byzantine-robust stochastic aggregation methods for distributed learning from heterogeneous datasets. In Association for the Ad- vancement of Artificial Intelligence, 2019.","title":"RSA: byzantine-robust stochastic aggregation methods for distributed learning from heterogeneous datasets","authors":"Liping Li, Wei Xu, Tianyi Chen, Georgios B Giannakis, Qing Ling","year":"2019","venue":"Association for the Advancement of Artificial Intelligence","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Li et al., 2022] Qinbin Li, Yiqun Diao, Quan Chen, and Bingsheng He. Federated learning on non-iid data silos: An experimental study. In IEEE International Conference on Data Engineering, 2022.","title":"Federated learning on non-iid data silos: An experimental study","authors":"Li","year":"2022","venue":"IEEE International Conference on Data Engineering","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Lin et al., 2019] Xi Lin, Hui-Ling Zhen, Zhenhua Li, Qing- Fu Zhang, and Sam Kwong. Pareto multi-task learning. In H. Wallach, H. Larochelle, A. Beygelzimer, F. d'Alché- Buc, E. Fox, and R. Garnett, editors, Advances in Neural Information Processing Systems, volume 32.","title":"Pareto multi-task learning","authors":"Xi Lin, Hui-Ling Zhen, Zhenhua Li, Qing-Fu Zhang, Sam Kwong","year":null,"venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[McMahan et al., 2017] Brendan McMahan, Eider Moore, Daniel Ramage, Seth Hampson, and Blaise Aguera y Ar- cas. Communication-Efficient Learning of Deep Networks from Decentralized Data. In Aarti Singh and Jerry Zhu, editors, Proceedings of the 20th International Conference on Artificial Intelligence and Statistics, volume 54 of Pro- ceedings of Machine Learning Research. PMLR, 20-22 Apr 2017.","title":"Communication-Efficient Learning of Deep Networks from Decentralized Data","authors":"Brendan Mcmahan, Eider Moore, Daniel Ramage, Seth Hampson, Blaise Aguera Y Arcas","year":"2017","venue":"Proceedings of the 20th International Conference on Artificial Intelligence and Statistics","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Mohri et al., 2019] Mehryar Mohri, Gary Sivek, and Ananda Theertha Suresh. Agnostic federated learning. In Kamalika Chaudhuri and Ruslan Salakhutdinov, edi- tors, Proceedings of the 36th International Conference on Machine Learning, volume 97 of Proceedings of Machine Learning Research. PMLR, 09-15 Jun 2019.","title":"Agnostic federated learning","authors":"Mehryar Mohri, Gary Sivek, Ananda Theertha, Suresh","year":"2019","venue":"Kamalika Chaudhuri and Ruslan Salakhutdinov, editors, Proceedings of the 36th International Conference on Machine Learning","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Netzer et al., 2011] Yuval Netzer, Tao Wang, Adam Coates, Alessandro Bissacco, Bo Wu, and Andrew Y. Ng. Reading digits in natural images with unsupervised feature learn- ing. In NeurIPS Workshop on Deep Learning and Unsu- pervised Feature Learning 2011, 2011.","title":"Reading digits in natural images with unsupervised feature learning","authors":"Yuval Netzer, Tao Wang, Adam Coates, Alessandro Bissacco, Bo Wu, Andrew Y Ng","year":"2011","venue":"NeurIPS Workshop on Deep Learning and Unsupervised Feature Learning","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Pillutla et al., 2022] Krishna Pillutla, Sham M. Kakade, and Zaid Harchaoui. Robust aggregation for federated learn- ing. In IEEE Transactions on Signal Processing, 2022.","title":"Robust aggregation for federated learning","authors":"Krishna Pillutla, M Sham, Zaid Kakade, Harchaoui","year":"2022","venue":"IEEE Transactions on Signal Processing","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Sener and Koltun, 2018] Ozan Sener and Vladlen Koltun. Multi-task learning as multi-objective optimization. In S. Bengio, H. Wallach, H. Larochelle, K. Grauman, N. Cesa-Bianchi, and R. Garnett, editors, Advances in Neural Information Processing Systems, volume 31. Cur- ran Associates, Inc., 2018.","title":"Multi-task learning as multi-objective optimization","authors":"Koltun ; ] Ozan Sener, Vladlen Sener, Koltun","year":"2018","venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":30,"raw":"[Shejwalkar and Houmansadr, 2021] Virat Shejwalkar and Amir Houmansadr. Manipulating the byzantine: Optimiz- ing model poisoning attacks and defenses for federated learning. In Network and Distributed Systems Security (NDSS) Symposium, 2021.","title":"Virat Shejwalkar and Amir Houmansadr. Manipulating the byzantine: Optimizing model poisoning attacks and defenses for federated learning","authors":"Houmansadr Shejwalkar","year":"2021","venue":"Network and Distributed Systems Security (NDSS) Symposium","doi":null,"url":null,"internal_url":null},{"position":31,"raw":"[Simonyan and Zisserman, 2015] Karen Simonyan and An- drew Zisserman. Very deep convolutional networks for large-scale image recognition. In International Confer- ence on Learning Representations, 2015.","title":"Very deep convolutional networks for large-scale image recognition","authors":"Karen Simonyan, Andrew Zisserman","year":"2015","venue":"International Conference on Learning Representations","doi":null,"url":null,"internal_url":null},{"position":32,"raw":"[Tang et al., 2022] Zhenheng Tang, Yonggang Zhang, Shao- huai Shi, Xin He, Bo Han, and Xiaowen Chu. Vir- tual homogeneity learning: Defending against data het- erogeneity in federated learning. In Kamalika Chaud- huri, Stefanie Jegelka, Le Song, Csaba Szepesvari, Gang Niu, and Sivan Sabato, editors, Proceedings of the 39th International Conference on Machine Learning, volume 162 of Proceedings of Machine Learning Research, pages 21111-21132. PMLR, 17-23 Jul 2022.","title":"Virtual homogeneity learning: Defending against data heterogeneity in federated learning","authors":"Tang","year":"2022","venue":"Proceedings of the 39th International Conference on Machine Learning","doi":null,"url":null,"internal_url":null},{"position":33,"raw":"[Xiao et al., 2017] Han Xiao, Kashif Rasul, and Roland Vollgraf. Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms. CoRR, abs/1708.07747, 2017.","title":"Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms","authors":"Xiao","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":34,"raw":"[Xie et al., 2020] Cong Xie, Oluwasanmi Koyejo, and In- dranil Gupta. Fall of empires: Breaking byzantine-tolerant sgd by inner product manipulation. In Proceedings of The 35th Uncertainty in Artificial Intelligence Conference, vol- ume 115 of Proceedings of Machine Learning Research. PMLR, 22-25 Jul 2020.","title":"Fall of empires: Breaking byzantine-tolerant sgd by inner product manipulation","authors":"Cong Xie, Oluwasanmi Koyejo, Indranil Gupta","year":"2020","venue":"Proceedings of The 35th Uncertainty in Artificial Intelligence Conference","doi":null,"url":null,"internal_url":null},{"position":35,"raw":"[Yin et al., 2018] Dong Yin, Yudong Chen, Ramchandran Kannan, and Peter Bartlett. Byzantine-robust distributed learning: Towards optimal statistical rates. In Jennifer Dy and Andreas Krause, editors, Proceedings of the 35th In- ternational Conference on Machine Learning, volume 80 of Proceedings of Machine Learning Research. PMLR, 10-15 Jul 2018.","title":"Byzantine-robust distributed learning: Towards optimal statistical rates","authors":"Dong Yin, Yudong Chen, Ramchandran Kannan, Peter Bartlett","year":"2018","venue":"Proceedings of the 35th International Conference on Machine Learning","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}