neural architecture search
This is an automated method for discovering optimal neural network architectures for specific tasks. By systematically evaluating different configurations, it aims to enhance model performance and efficiency without requiring extensive manual tuning.
- GC4NC: A Benchmark Framework for Graph Condensation on Node Classification with New Insights
- LoMix: Learnable Weighted Multi-Scale Logits Mixing for Medical Image Segmentation
- Omnipresent Yet Overlooked: Heat Kernels in Combinatorial Bayesian Optimization
- Per-Architecture Training-Free Metric Optimization for Neural Architecture Search
- Revolutionizing Training-Free NAS: Towards Efficient Automatic Proxy Discovery via Large Language Models
- Searching Efficient Semantic Segmentation Architectures via Dynamic Path Selection
- TF-MAS: Training-free Mamba2 Architecture Search