Federated learning

Better Methods and Theory for Federated Learning: Compression, Client Selection, and Heterogeneity

-

B5 L5 R5209

Federated learning (FL) is an emerging machine learning paradigm involving multiple clients, e.g., mobile phone devices, with an incentive to collaborate in solving a machine learning problem coordinated by a central server. FL was proposed in 2016 by Konecny et al. and McMahan et al. as a viable privacy-preserving alternative to traditional centralized machine learning since, by construction, the training data points are decentralized and never transferred by the clients to a central server. Therefore, to a certain degree, FL mitigates the privacy risks associated with centralized data collection. Unfortunately, optimization for FL faces several specific issues that centralized optimization usually does not need to handle. In this thesis, we identify several of these challenges and propose new methods and algorithms to address them, with the ultimate goal of enabling practical FL solutions supported with mathematically rigorous guarantees.

EF-BV: distributed optimization with compressed communication

-

B9 L2 R2322 H1

In the big data era, it is necessary to rely on distributed computing. For distributed optimization and learning tasks, in particular in the modern paradigm of federated learning, specific challenges arise, such as decentralized data storage. Communication between the parallel machines and the orchestrating distant server is necessary but slow. To address this main bottleneck, a natural strategy is to compress the communicated vectors. I will present EF-BV, a new algorithm which converges linearly to an exact solution, with a large class of deterministic or random, biased or unbiased compressors.

Making federated learning practical and confidential

Dr. Ruichuan Chen, Distinguished Member of Technical Staff and a Tech Lead, Nokia Bell Labs

-

B4 L5 R5209

Federated learning (FL) is increasingly deployed among multiple clients to train a shared model over decentralized data. To address the privacy concerns, FL systems need to protect the clients' data from being revealed during training, and also control data leakage through trained models when exposed to untrusted domains. However, existing FL systems (with distributed differential privacy) work impractically in the presence of client dropout, resulting in either poor privacy guarantees or degraded training accuracy. In addition, existing FL systems focus on safeguarding the privacy of training data, but not on protecting the confidentiality of the models being trained, which are increasingly of high business value. In this talk, I will present two pieces of our recent work that aim to address these aforementioned issues.

FADO: A Federated Learning Attack and Defense Orchestrator

-

B5 L5 R5209

Federated Learning (FL) is a distributed machine learning approach that allows multiple parties to train a model collaboratively without sharing sensitive data.

Resistance is Futile: Balancing Opportunities and Security Challenges of Federated Learning

-

B4 L5 R5220

The rapid growth of Artificial Intelligence (AI) and Deep Learning mirrors an infectious phenomenon. While AI systems promise diverse applications and benefits, they bear substantial security and privacy risks. Indeed, AI represents a goldmine for the security and privacy research domain.

Ziwu Liu

M.S. Student, Computer Science
My passion for the evolving landscape of technology has driven me to focus on the intricate interplay between resilient computing, federated learning, edge computing, and machine learning systems. My research aims to seamlessly integrate edge computing with federated learning to create resilient and adaptive machine learning systems that can operate effectively even in resource-constrained settings.

Rafal Szlendak

Rafał Szlendak is an MS/PhD student in the AMCS program under the supervision of Prof. Peter Richtarik. His biggest passions are research, weightlifting and travelling. His career path is motivated by his endless passion for self-development and discovery. Education and Early Career Rafał Szlendak has obtained his Bachelor's degree in Mathematics and Statistics from University of Warwick Research Interest In his research Rafał focuses on stochastic gradient descent algorithms for large-scale machine learning problems. His primary goal is developing state-of-the-art methods with sound

Norah S. Alballa

Ph.D. Student, Computer Science
Education Profile Ph.D. in Computer Science, CEMSE, KAUST, 2022-present M.S. in Computer Science, Portland State University, 2016 B.S. in Information Technology, King Saud University (KSU), 2010

Laurent Condat

I am a senior researcher working in optimization, in the group of Prof. Peter Richtárik.

Konstantin Burlachenko

Ph.D. Student, Computer Science
Konstantin Burlachenko is a Ph.D. candidate at the KAUST Optimization and Machine Learning Lab under the supervision of Professor Peter Richtarik.

Grigorii Malinovskii

Ph.D. Student, Computer Science
I am a PhD student in Applied Mathematics and Computational Science at King Abdullah University of Science and Technology, advised by Prof. Peter Richtárik, working on optimization theory and its applications in Machine Learning. My hobbies include drums, fitness, martial arts, football, and basketball.

Egor Shulgin

Ph.D. Student, Computer Science
Egor Shulgin is a CS student in the MS/Ph.D. program under the supervision of Professor Peter Richtarik at the Visual Computing Center (VCC) at the King Abdullah University of Science and Technology (KAUST), Saudi Arabia. Education and Early Career Egor obtained his bachelor's degree in Applied Mathematics, Physics, and Computer Science from the Moscow Institute of Physics and Technology (MIPT) in 2019. He first came to KAUST for a research visit in 2019. Later, he joined Prof. Peter Richtarik's group as a MS/Ph.D. student. In 2021 Egor did a research internship with Distributed AI team at the

Boris Radovic

Boris Radovic is a Ph.D. candidate in the computer science program, under the supervision of Professors Marco Canini and Veljko Pejović, and is a member of the Software-defined Advanced Networked and Distributed Systems (SANDS) Lab. He holds a Bachelor of Science (B.Sc.) in Computer Science and a Master of Science (M.Sc.) in Data Science, both obtained from the Faculty of Computer and Information Science at the University of Ljubljana, Slovenia. During his academic journey, Boris gained experience across diverse computer science domains through his long-term collaboration with the Laboratory