Practical Method
Practical methods in machine learning and related fields are currently focused on improving efficiency, accuracy, and generalizability of existing algorithms and models. Research emphasizes developing faster solvers for optimization problems (e.g., using parallel-in-time methods and novel optimizers like the generalized Newton's method), enhancing model robustness through techniques such as low-rank approximations and prompt portfolios, and creating more reliable uncertainty quantification methods. These advancements are crucial for deploying machine learning models in resource-constrained environments and for building more trustworthy and explainable AI systems across diverse applications.
Papers
FedDP: Privacy-preserving method based on federated learning for histopathology image segmentation
Liangrui Pan, Mao Huang, Lian Wang, Pinle Qin, Shaoliang Peng
Unlearning in- vs. out-of-distribution data in LLMs under gradient-based method
Teodora Baluta, Pascal Lamblin, Daniel Tarlow, Fabian Pedregosa, Gintare Karolina Dziugaite
RS-MOCO: A deep learning-based topology-preserving image registration method for cardiac T1 mapping
Chiyi Huang, Longwei Sun, Dong Liang, Haifeng Liang, Hongwu Zeng, Yanjie Zhu
UmambaTSF: A U-shaped Multi-Scale Long-Term Time Series Forecasting Method Using Mamba
Li Wu, Wenbin Pei, Jiulong Jiao, Qiang Zhang
Dissecting embedding method: learning higher-order structures from data
Liubov Tupikina (UPD5, LPI), Kathuria Hritika (LPI)
Copula-Linked Parallel ICA: A Method for Coupling Structural and Functional MRI brain Networks
Oktay Agcaoglu, Rogers F. Silva, Deniz Alacam, Sergey Plis, Tulay Adali, Vince Calhoun (for the Alzheimers Disease Neuroimaging Initiative)