3 papers
cs.CV2025
Active Learning from Scene Embeddings for End-to-End Autonomous Driving
Wenhao Jiang, Duo Li, Menghan Hu +3
In the field of autonomous driving, end-to-end deep learning models show great potential by learning driving decisions directly from sensor data. However, training these models req…
cs.LG2024
Few-Shot Class-Incremental Learning with Prior Knowledge
Wenhao Jiang, Duo Li, Menghan Hu +3
To tackle the issues of catastrophic forgetting and overfitting in few-shot class-incremental learning (FSCIL), previous work has primarily concentrated on preserving the memory of…
cs.CV2024
Uncertainty-aware Sampling for Long-tailed Semi-supervised Learning
Kuo Yang, Duo Li, Menghan Hu +3
For semi-supervised learning with imbalance classes, the long-tailed distribution of data will increase the model prediction bias toward dominant classes, undermining performance o…