Publications

2024

Cohort Squeeze: Beyond a Single Communication Round per Cohort in Cross-Device Federated Learning
Kai Yi, Timur Kharisov, Igor Sokolov, Peter Richtárik
NeurIPS 2024 FL@FM Workshop (Oral)
arXiv, 2024. [arXiv]
Prune at the Clients, Not the Server: Accelerated Sparse Training in Federated Learning
Georg Meinhardt, Kai Yi, Laurent Condat, Peter Richtárik
arXiv, 2024. [arXiv]
PV-Tuning: Beyond Straight-Through Estimation for Extreme LLM Compression
Vladimir Malinovskii, Denis Mazur, Ivan Ilin, Denis Kuznedelev, Konstantin Burlachenko, Kai Yi, Dan Alistarh, Peter Richtarik
NeurIPS (Oral), 2024. [arXiv] [code] [HF Models]
FedComLoc: Communication-Efficient Distributed Training of Sparse and Quantized Models
Kai Yi, Georg Meinhardt, Laurent Condat, Peter Richtárik
arXiv, 2024. [arXiv] [code]
FedP3: Federated Personalized and Privacy-friendly Network Pruning under Model Heterogeneity
Kai Yi, Nidham Gazagnadou, Peter Richtárik, Lingjuan Lyu
ICLR, 2024. [paper] [code]

Domain-Aware Continual Zero-Shot Learning
Kai Yi, Mohamed Elhoseiny
ICCV Workshop OOD-CV, 2023. [paper] [project]
Continual Zero-Shot Learning through Semantically Guided Generative Random Walks
Wenxuan Zhang, Paul Janson, Kai Yi, Ivan Skorokhodov, Mohamed Elhoseiny
ICCV, 2023. [paper] [code]
Explicit Personalization and Local Training: Double Communication Acceleration in Federated Learning
Kai Yi, Laurent Condat, Peter Richtárik
arXiv, 2023. [paper] [code]

Variance Reduced ProxSkip: Algorithm, Theory and Application to Federated Learning
Grigory Malinovsky, Kai Yi, Peter Richtárik
NeurIPS, 2022. [paper] [code]
Exploring Hierarchical Graph Representation for Large-Scale Zero-Shot Image Classification
Kai Yi, Xiaoqian Shen, Yunhao Gou, Mohamed Elhoseiny
ECCV, 2022. [paper] [project] [code]
EF-BV: A Unified Theory of Error Feedback and Variance Reduction Mechanisms for Biased and Unbiased Compression in Distributed Optimization
Laurent Condat, Kai Yi, Peter Richtárik
NeurIPS, 2022. [paper] [code]
Language-Guided Imaginative Walks: Generative Random Walk Deviation Loss for Unseen Class Recognition using Text
Kai Yi, Divyansh Jha, Ivan Skorokhodov, Mohamed Elhoseiny
CVPR, 2022, L3D-IVU Workshop (Short Paper). [paper] [code]
Creative Walk Adversarial Networks: Novel Art Generation with Probabilistic Random Walk Deviation from Style Norms
Divyansh Jha*, Kai Yi, Ivan Skorokhodov, Mohamed Elhoseiny*
ICCC, 2022. [project] [paper] [code]
VisualGPT: Data-efficient Image Captioning by Balancing Visual Input and Linguistic Knowledge from Pretraining
Jun Chen, Han Guo, Kai Yi, Boyang Li, Mohamed Elhoseiny
CVPR, 2022. [arXiv] [CVF] [code]
CIZSL++: Creativity Inspired Generative Zero-Shot Learning
Mohamed Elhoseiny, Kai Yi, Mohamed Elfeki
T-PAMI Major Revision. [paper] [code].

Domain-Aware Continual Zero-Shot Learning
Kai Yi
Master Thesis, 2021. [thesis]
Imaginative Walks: Generative Random Walk Deviation Loss for Improved Unseen Learning Representation
Divyansh Jha*, Kai Yi*, Ivan Skorokhodov, Mohamed Elhoseiny
arXiv, 2021. [project page] [paper] [code]
Disentangling semantic features of macromolecules in Cryo-Electron Tomography
Kai Yi, Jianye Pang, Yungeng Zhang, Xiangrui Zeng, Min Xu
arXiv, 2021. [paper]
Unsupervised Domain Alignment based Open Set Structural Recognition of Macromolecules Captured by Cryo-Electron Tomography
Yuchen Zeng, Gregory Howe, Kai Yi, Xiangrui Zeng, Jing Zhang, Yi-Wei Chang, Min Xu
ICIP 2021. [paper]

Experimental Analysis of Legendre Decomposition in Machine Learning
Jianye Pang, Kai Yi, Wanguang Yin, Min Xu
Technical Report, 2020. [arXiv]
Feature Selective Small Object Detection via Knowledge-based Recurrent Attentive Neural Network
Kai Yi, Zhiqiang Jian, Shitao Chen, Nanning Zheng
Technical Report, 2019. [paper]
Affine LBG for Codebook Training of Univariate Linear Approximation
Tiannan Dong, Jianji Wang, Meng Yang, Yi Kai, Nanning Zheng
GlobalSIP, 2018. [paper]
Cognition-Based Deep Learning: Progresses and Perspectives
Kai Yi, Shitao Chen, Yu Chen, Chao Xia, Nanning Zheng
AIAI, 2018. [paper]