Workshop
International Workshop on Federated Foundation Models in Conjunction with NeurIPS 2024 (FL@FM-NeurIPS'24)
Sai Praneeth Karimireddy ยท Xiaoxiao Li ยท Songtao Lu ยท Stacy Patterson ยท Pascal Poupart ยท Han Yu
East Meeting Room 8, 15
Sun 15 Dec, 8:15 a.m. PST
The rise of foundation models (FMs) amplifies the importance and relevance of federated learning (FL) as a crucial research direction. With FMs becoming the norm in machine learning development, the focus shifts from model architecture design to tackling the issues surrounding privacy-preserving and distributed learning. Advancements in FL methods have the potential to unlock the use of FMs, enabling efficient and scalable training while safeguarding sensitive data. With this in mind, we invite original research contributions, position papers, and work-in-progress reports on various aspects of federated learning in the era of foundation models. Since the emergence of foundation models has been a relatively recent phenomenon, their full impact on federated learning has not yet been well explored or understood. We hope to provide a platform to facilitate interaction among students, scholars, and industry professionals from around the world to discuss the latest advancements, share insights, and identify future directions in this exciting field. This workshop aims to bring together academic researchers and industry practitioners to address open issues in this interdisciplinary research area. For industry participants, we intend to create a forum to communicate problems are practically relevant. For academic participants, we hope to make it easier to become productive in this area. The workshop will focus on the theme of combining FL with FM to open up opportunities to address new challenges.
Schedule
Sun 8:15 a.m. - 8:20 a.m.
|
Opening Remarks
(
Intro
)
>
SlidesLive Video |
๐ |
Sun 8:20 a.m. - 9:00 a.m.
|
Federated Large Language Models and Their Applications, by Qiang Yang
(
Keynote Talk
)
>
SlidesLive Video |
Qiang Yang ๐ |
Sun 9:00 a.m. - 9:07 a.m.
|
Alex Iacob, Lorenzo Sani, Bill Marino, Preslav Aleksandrov, William F. Shen & Nicholas Donald Lane. Worldwide Federated Training of Language Models
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 9:07 a.m. - 9:14 a.m.
|
Alexander Bienstock, Antigoni Polychroniadou & Ujjwal Kumar. Distributed Matrix Mechanism for Federated Learning
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 9:14 a.m. - 9:21 a.m.
|
Christian Internรฒ, Elena Raponi, Niki van Stein, Thomas Bรคck, Markus Olhofer, Yaochu Jin & Barbara Hammer. Adaptive Model Hybrid Pruning in Federated Learning through Loss Exploration
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 9:21 a.m. - 9:28 a.m.
|
Filip Granqvist, Congzheng Song, รine Cahill, Rogier van Dalen, Martin Pelikan, Yi Sheng Chan, Xiaojun Feng, Natarajan Krishnaswami, Vojta J & Mona Chitnis. pfl-research: Simulation Framework for Accelerating Research in Private Federated Learning
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 9:30 a.m. - 10:00 a.m.
|
Coffee Break
|
๐ |
Sun 10:00 a.m. - 10:07 a.m.
|
Harish Karthikeyan & Antigoni Polychroniadou. OPA: One-shot Private Aggregation with Single Client Interaction and its Applications to Federated Learning
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 10:07 a.m. - 10:14 a.m.
|
Kai Yi, Timur Kharisov, Igor Sokolov & Peter Richtรกrik. Cohort Squeeze: Beyond a Single Communication Round per Cohort in Cross-Device Federated Learning
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 10:14 a.m. - 10:21 a.m.
|
Lei Shen, Zhenheng Tang, Lijun Wu, Yonggang Zhang, Xiaowen Chu, Tao Qin & Bo Han. Hot Pluggable Federated Learning
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 10:21 a.m. - 10:28 a.m.
|
Lorenzo Sani, Alex Iacob, Zeyu Cao, Bill Marino, Yan Gao, Tomas Paulik, Wanru Zhao, William F. Shen, Preslav Aleksandrov, Xinchi Qiu & Nicholas Donald Lane. The Future of Large Language Model Pre-training is Federated
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 10:30 a.m. - 11:00 a.m.
|
The first AGI will be Federated, by Nicholas Lane
(
Keynote Talk
)
>
SlidesLive Video |
Nicholas Lane ๐ |
Sun 11:00 a.m. - 11:07 a.m.
|
Lu Li, Tianyu Zhang, Zhiqi Bu, Suyuchen Wang, Huan He, Jie Fu, Yonghui Wu, Jiang Bian, Yong Chen & Yoshua Bengio. MAP: Model Merging with Amortized Pareto Front Using Limited Computation
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 11:07 a.m. - 11:14 a.m.
|
Mariel Werner, Sai Praneeth Karimireddy & Michael Jordan. Defection-Free Collaboration between Competitors in a Learning System ( Paper Presentation ) > link | ๐ |
Sun 11:14 a.m. - 11:21 a.m.
|
Muxing Wang, Pengkun Yang & Lili Su. On the Convergence Rates of Federated Q-Learning across Heterogeneous Environments
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 11:21 a.m. - 11:28 a.m.
|
Rui Ye, Jingyi Chai, Xiangrui Liu, Yaodong Yang, Yanfeng Wang & Siheng Chen. Emerging Safety Attack and Defense in Federated Instruction Tuning of Large Language Models ( Paper Presentation ) > link | ๐ |
Sun 11:28 a.m. - 11:35 a.m.
|
Rui Ye, Rui Ge, Fengting Yuchi, Jingyi Chai, Yanfeng Wang & Siheng Chen. Leveraging Unstructured Text Data for Federated Instruction Tuning of Large Language Models ( Paper Presentation ) > link | ๐ |
Sun 11:35 a.m. - 11:42 a.m.
|
Rui Ye, Xinyu Zhu, Jingyi Chai, Lingjuan Lyu, Chen Xie, Yanfeng Wang & Siheng Chen. Federated Learning with Generative Content ( Paper Presentation ) > link | ๐ |
Sun 11:42 a.m. - 11:49 a.m.
|
Sergio Zaera Mata & Roberto Gรณmez-Espinosa Martรญn. The SynapticCity Phenomenon: When All Foundation Models Marry Federated Learning and Blockchain
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 11:49 a.m. - 11:56 a.m.
|
Steffen Schotthรถfer & M. Paul Laiu. Federated Dynamical Low-Rank Training with Global Loss Convergence Guarantees
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 11:56 a.m. - 12:03 p.m.
|
Sunny Gupta & Amit Sethi. FedStein: Enhancing Multi-Domain Federated Learning Through James-Stein Estimator
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 12:03 p.m. - 12:10 p.m.
|
Tao Yu, Congzheng Song, Jianyu Wang & Mona Chitnis. Momentum Approximation in Asynchronous Private Federated Learning
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 12:30 p.m. - 2:00 p.m.
|
Lunch Break
|
๐ |
Sun 2:00 p.m. - 2:30 p.m.
|
Transforming Multicenter Neurology Trials with Federated Learning: A New Era of Collaborative Medicine, by Martin J. McKeown
(
Keynote Talk
)
>
SlidesLive Video |
๐ |
Sun 2:30 p.m. - 3:00 p.m.
|
Federated Optimization Beyond Standard Empirical Risk Minimization, by Gauri Joshi
(
Keynote Talk
)
>
SlidesLive Video |
Gauri Joshi ๐ |
Sun 3:00 p.m. - 3:30 p.m.
|
Coffee Break
|
๐ |
Sun 3:30 p.m. - 4:00 p.m.
|
Machine Learning from Imbalanced Data Sources, by Shiqiang Wang
(
Keynote Talk
)
>
SlidesLive Video |
Shiqiang Wang ๐ |
Sun 4:00 p.m. - 4:07 p.m.
|
Vasileios Tsouvalas, Samaneh Mohammadi, Ali Balador, Tanir รzรงelebi, Francesco Flammini & Nirvana Meratnia. EncCluster: Bringing Functional Encryption in Federated Foundational Models
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 4:07 p.m. - 4:14 p.m.
|
Wang Lu, Hao Yu, Jindong Wang, Damien Teney, Haohan Wang, Yao Zhu, Yiqiang Chen, Qiang Yang, Xing Xie & Xiangyang Ji. ZOOPFL: Exploring Black-box Foundation Models for Personalized Federated Learning
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 4:14 p.m. - 4:21 p.m.
|
Xianjie Guo, Liping Yi, Xiaohu Wu, Kui Yu & Gang Wang. Enhancing Causal Discovery in Federated Settings with Limited Local Samples
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 4:21 p.m. - 4:28 p.m.
|
Xiaochun Niu, Lili Su, Jiaming Xu & Pengkun Yang. Collaborative Learning with Shared Linear Representation: Statistical Rates and Optimal Algorithms
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 4:28 p.m. - 4:35 p.m.
|
Yao Shu, Wenyang Hu, See-Kiong Ng, Bryan Kian Hsiang Low & Fei Richard Yu. Ferret: Federated Full-Parameter Tuning at Scale for Large Language Models
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 4:35 p.m. - 4:42 p.m.
|
Zexi Li, Jie Lin, Zhiqi Li, Didi Zhu, Rui Ye, Tao Shen, Tao Lin & Chao Wu. Improving Group Connectivity for Generalization of Federated Deep Learning ( Paper Presentation ) > link | ๐ |
Sun 4:42 p.m. - 4:49 p.m.
|
Zhe Li, Bicheng Ying, Zidong Liu, Chaosheng Dong & Haibo Yang. DeComFL: Federated Learning with Dimension-Free Communication
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 4:49 p.m. - 4:56 p.m.
|
Zhilong Li, Xiaohu Wu, Xiaoli Tang, Tiantian He, Yew-Soon Ong, Mengmeng Chen, Qiqi Liu, Qicheng Lao, Xiaoxiao Li & Han Yu. Benchmarking Data Heterogeneity Evaluation Approaches for Personalized Federated Learning
(
Paper Presentation
)
>
link
SlidesLive Video |
๐ |
Sun 5:00 p.m. - 5:15 p.m.
|
Award Ceremony
(
Closing
)
>
SlidesLive Video |
๐ |
-
|
ZOOPFL: Exploring Black-box Foundation Models for Personalized Federated Learning ( Oral ) > link | Wang Lu ยท ๆตฉ ไฝ ยท Jindong Wang ยท Damien Teney ยท Haohan Wang ยท Yao Zhu ยท Yiqiang Chen ยท Qiang Yang ยท Xing Xie ยท Xiangyang Ji ๐ |
-
|
Improving Group Connectivity for Generalization of Federated Deep Learning ( Oral ) > link | Zexi Li ยท Jie Lin ยท Zhiqi Li ยท Didi Zhu ยท Rui Ye ยท Tao Shen ยท Tao Lin ยท Chao Wu ๐ |
-
|
FedStein: Enhancing Multi-Domain Federated Learning Through James-Stein Estimator ( Oral ) > link | Sunny Kumar Gupta ยท Amit Sethi ๐ |
-
|
OPA: One-shot Private Aggregation with Single Client Interaction and its Applications to Federated Learning ( Oral ) > link | Harish Karthikeyan ยท Antigoni Polychroniadou ๐ |
-
|
Leveraging Unstructured Text Data for Federated Instruction Tuning of Large Language Models ( Oral ) > link | Rui Ye ยท Rui Ge ยท Fengting Yuchi ยท Jingyi Chai ยท Yanfeng Wang ยท Siheng Chen ๐ |
-
|
Momentum Approximation in Asynchronous Private Federated Learning ( Oral ) > link | Tao Yu ยท Congzheng Song ยท Jianyu Wang ยท Mona Chitnis ๐ |
-
|
Collaborative Learning with Shared Linear Representations: Statistical Rates and Optimal Algorithms ( Oral ) > link | Xiaochun Niu ยท Lili Su ยท Jiaming Xu ยท Pengkun Yang ๐ |
-
|
The Future of Large Language Model Pre-training is Federated ( Oral ) > link |
11 presentersLorenzo Sani ยท Alexandru-Andrei Iacob ยท Zeyu Cao ยท Bill Marino ยท Yan Gao ยท Tomas Paulik ยท Wanru Zhao ยท William Shen ยท Preslav Aleksandrov ยท Xinchi Qiu ยท Nicholas Lane |
-
|
Federated Learning with Generative Content ( Oral ) > link | Rui Ye ยท Xinyu Zhu ยท Jingyi Chai ยท Lingjuan Lyu ยท Chen Xie ยท Yanfeng Wang ยท Siheng Chen ๐ |
-
|
Worldwide Federated Training of Language Models ( Oral ) > link | Alexandru-Andrei Iacob ยท Lorenzo Sani ยท Bill Marino ยท Preslav Aleksandrov ยท William Shen ยท Nicholas Lane ๐ |
-
|
DeComFL: Federated Learning with Dimension-Free Communication ( Oral ) > link | Zhe Li ยท Bicheng Ying ยท Zidong Liu ยท Chaosheng Dong ยท Haibo Yang ๐ |
-
|
Benchmarking Data Heterogeneity Evaluation Approaches for Personalized Federated Learning ( Oral ) > link | Zhilong Li ยท Xiaohu Wu ยท Xiaoli Tang ยท Tiantian He ยท Yew Soon Ong ยท Mengmeng Chen ยท QIQI LIU ยท Qicheng Lao ยท Han Yu ๐ |
-
|
Ferret: Federated Full-Parameter Tuning at Scale for Large Language Models ( Oral ) > link | Yao Shu ยท Wenyang Hu ยท See-Kiong Ng ยท Bryan Kian Hsiang Low ยท Fei Yu ๐ |
-
|
The SynapticCity Phenomenon: When All Foundation Models Marry Federated Learning and Blockchain ( Oral ) > link | Sergio Zaera Mata ยท Roberto Gรณmez-Espinosa Martรญn ๐ |
-
|
Adaptive Hybrid Model Pruning in Federated Learning through Loss Exploration ( Oral ) > link | Christian Internรฒ ยท Elena Raponi ยท Niki van Stein ยท Thomas Bรคck ยท Markus Olhofer ยท Yaochu Jin ยท Barbara Hammer ๐ |
-
|
Federated Dynamical Low-Rank Training with Global Loss Convergence Guarantees ( Oral ) > link | Steffen Schotthรถfer ยท M. Laiu ๐ |
-
|
MAP: Model Merging with Amortized Pareto Front Using Limited Computation ( Oral ) > link | Lu Li ยท Tianyu Zhang ยท Zhiqi Bu ยท Suyuchen Wang ยท Huan He ยท Jie Fu ยท Yonghui Wu ยท Jiang Bian ยท Yong Chen ยท Yoshua Bengio ๐ |
-
|
Enhancing Causal Discovery in Federated Settings with Limited Local Samples ( Oral ) > link | Xianjie Guo ยท Liping Yi ยท Xiaohu Wu ยท Kui Yu ยท Gang Wang ๐ |
-
|
DMM: Distributed Matrix Mechanism for Differentially-Private Federated Learning using Packed Secret Sharing ( Oral ) > link | Alexander Bienstock ยท Antigoni Polychroniadou ยท Ujjwal Kumar ๐ |
-
|
Hot Pluggable Federated Learning ( Oral ) > link | Lei SHEN ยท Zhenheng Tang ยท Lijun Wu ยท Yonggang Zhang ยท Xiaowen Chu ยท Tao Qin ยท Bo Han ๐ |
-
|
On the Convergence Rates of Federated Q-Learning across Heterogeneous Environments ( Oral ) > link | Muxing Wang ยท Pengkun Yang ยท Lili Su ๐ |
-
|
Cohort Squeeze: Beyond a Single Communication Round per Cohort in Cross-Device Federated Learning ( Oral ) > link | Kai Yi ยท Timur Kharisov ยท Igor Sokolov ยท Peter Richtarik ๐ |
-
|
EncCluster: Bringing Functional Encryption in Federated Foundational Models ( Oral ) > link | Vasileios Tsouvalas ยท Samaneh Mohammadi ยท Ali Balador ยท Tanir รzรงelebi ยท Francesco Flammini ยท Nirvana Meratnia ๐ |
-
|
: simulation framework for accelerating research in Private Federated Learning ( Oral ) > link | Filip Granqvist ยท Congzheng Song ยท รine Cahill ยท Rogier van Dalen ยท Martin Pelikan ยท Yi Sheng Chan ยท Xiaojun Feng ยท Natarajan Krishnaswami ยท Vojta Jina ยท Mona Chitnis ๐ |
-
|
Defection-Free Collaboration between Competitors in a Learning System ( Oral ) > link | Mariel Werner ยท Sai Praneeth Karimireddy ยท Michael Jordan ๐ |
-
|
Emerging Safety Attack and Defense in Federated Instruction Tuning of Large Language Models ( Oral ) > link | Rui Ye ยท Jingyi Chai ยท Xiangrui Liu ยท Yaodong Yang ยท Yanfeng Wang ยท Siheng Chen ๐ |