6 papers
Quantifying the Privacy Implications of High-Fidelity Synthetic Network Traffic
Van Tran, Shinan Liu, Tian Li +1
To address the scarcity and privacy concerns of network traffic data, various generative models have been developed to produce synthetic traffic. However, synthetic traffic is not…
Efficient Adaptive Federated Optimization
Su Hyeong Lee, Sidharth Sharma, Manzil Zaheer +1
Adaptive optimization is critical in federated learning, where enabling adaptivity on both the server and client sides has proven essential for achieving optimal performance. Howev…
Efficient Distributed Optimization under Heavy-Tailed Noise
Su Hyeong Lee, Manzil Zaheer, Tian Li
Distributed optimization has become the default training paradigm in modern machine learning due to the growing scale of models and datasets. To mitigate communication overhead, lo…
Differentially Private Federated Clustering with Random Rebalancing
Xiyuan Yang, Shengyuan Hu, Soyeon Kim +1
Federated clustering aims to group similar clients into clusters and produce one model for each cluster. Such a personalization approach typically improves model performance compar…
Triton-distributed: Programming Overlapping Kernels on Distributed AI Systems with the Triton Compiler
Size Zheng, Wenlei Bao, Qi Hou +19
In this report, we propose Triton-distributed, an extension of existing Triton compiler, to overcome the programming challenges in distributed AI systems. Triton-distributed is the…
Topology-Aware Knowledge Propagation in Decentralized Learning
Mansi Sakarvadia, Nathaniel Hudson, Tian Li +2
Decentralized learning enables collaborative training of models across naturally distributed data without centralized coordination or maintenance of a global model. Instead, device…