Why Federated Optimization Fails to Achieve Perfect Fitting? A Theoretical Perspective on Client-Side Optima
"理论分析" 相关文章
On the Optimal Construction of Unbiased Gradient Estimators for Zeroth-Order Optimization
Learning under Quantization for High-Dimensional Linear Regression
Budget-constrained Active Learning to Effectively De-censor Survival Data
Critical attention scaling in long-context transformers
On the Limitations and Capabilities of Position Embeddings for Length Generalization
Calibration Meets Reality: Making Machine Learning Predictions Trustworthy
A Theoretical Analysis of Discrete Flow Matching Generative Models
Adam的Update RMS为何总是0.2？噪声模拟到理论近似全讲透
求道之人，不问寒暑（九）
Towards High-Order Mean Flow Generative Models: Feasibility, Expressivity, and Provably Efficient Criteria
A Lower Bound for the Number of Linear Regions of Ternary ReLU Regression Neural Networks
Mathematical Introduction to Deep Learning: Methods, Implementations, and Theory
What Makes Local Updates Effective: The Role of Data Heterogeneity and Smoothness
再次颠覆学界想象，何恺明发表新作：扩散模型不一定需要噪声条件
文本-图像全局对比对齐与 Token-Patch 级别的局部对齐
