conftrace_

Mingzhen He

4 papers · 2023–2025 · 3 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🌉 Interdisciplinary Bridge · Bronze

Venues

NeurIPS (2) ICLR (1) ICML (1)

Top co-authors

Xiaolin Huang (4) Ruikai Yang (3) Hanling Tian (2) jie Yang (1) Kexin Lv (1) Kun Fang (1) FAN He (1) Youmei Qiu (1) Zhehao Huang (1) Zhengbao He (1)

Research topics

Machine Learning (4) Application Areas (2) Core Methods (2) Deep Learning (1) Mathematics & Optimization (1) Architectures (1) Learning Types (1) Mathematics (1) Optimization & Theory (1) Techniques (1)

Keywords

principal component analysis (1) neural network optimization (1) dimension reduction (1) kernel pca (1) reconstruction error (1) directed graph (1) non-linear mapping (1) parameter recovery (1) linear complexity (1) graph transformer (1) out-of-distribution detection (1) training dynamics (1) self attention (1) data reconstruction (1) diffusion map (1) nonlinear dimension reduction (1) universal approximator (1) dual representation (1) asymmetric kernel (1) magnetic transform (1)

Papers

Primphormer: Efficient Graph Transformers with Primal Representations ICML 2025 Simulating Training Dynamics to Reconstruct Training Data from Deep Neural Networks ICLR 2025 Kernel PCA for Out-of-Distribution Detection NeurIPS 2024 Diffusion Representation for Asymmetric Kernels via Magnetic Transform NeurIPS 2023