👋 About Me
Hello 🙋, I'm Haokun Ren, a master's student in Artificial Intelligence at Chongqing University of Posts and Telecommunications, advised by Professors Ke Liu and Ye Wang. My research interests include brain–computer interfaces, representation learning, and contrastive learning. You can reach me at j4ck1e2000@gmail.com.
🧠 Connecting neural signals and visual representations
I study how non-invasive neural signals reveal visual representations, with a focus on EEG/MEG brain-to-image retrieval, alignment between neural responses and visual features, and learning across the visual hierarchy.
📚 Publications
The Visual Target Matters: Learning across the Visual Hierarchy for Brain-to-Image Retrieval
NeuroGlyph learns a trial-independent visual target from multiple depths of a frozen vision backbone, combining factor-specific subspaces with image-conditioned depth allocation.
Adaptive Cortically Constrained EEG-Vision Alignment for Zero-Shot Brain-to-Image Retrieval
Maps EEG responses to predefined ROI-level source-pattern representations and adaptively weights visual supervision using sample-level alignment evidence for zero-shot brain-to-image retrieval.
🤝 Collaborators
- Maocai Dai (戴茂才)
- Xuyang Zhou (周绪洋)
📖 Educations
Master's student in Artificial Intelligence at Chongqing University of Posts and Telecommunications.