A biweekly reading group to discuss recent or classical papers on transfer learning methodologies and their applications on medical tasks. Other interesting theory topics are also welcomed.

Time: Monday 10:30 a.m. (GMT+8)

• Presenter: Wu Yanru
• Paper: Auxiliary Task Reweighting for Minimum-data Learning (Shi, NeurIPS 2020)
• Slides: auxiliary_task_reweighting.pptx

• Presenter: Duan Shutong
• A Brief Research in Semantic Segmentation with Multisource
• Slides:brief_literature_review.pptx

• Presenter: Zhao Zixi
• Paper: Test-time Adaptation in the Dynamic World with Compound Domain Knowledge Management (Song, IEEE Robotics and Automation Letters 2022)
• Slides: 5.10_pre.pptx

• Presenter: Lai Jiahao
• Paper: Variational Continual Learning (Nguyen, ICLR 2018)
• Slides: 5.24组会.pdf

• Presenter: Wu Yanru
• Paper: A Geometric Analysis of Neural Collapse with Unconstrained Features (Zhu and Ding, NeurIPS 2021)
• Slides: a_geometric_analysis_of_neural_collapse_with.pptx

• Presenter: Zhao Zixi
• Paper: Mad Max: Affine Spline Insights into Deep Learning (Balestriero, expanded from ICML 2018)
• Slides: 7.5_pre.pptx

• Presenter: Dong Caixia
• Paper: High-quality coronary artery segmentation via fuzzy logic modeling coupled with dynamic graph convolutional network
• Slides: 冠脉分割20230802dcx.pdf

• Presenter: Duan Shutong; Zhang Enming
• Paper: Discriminability and Transferability Estimation: A Bayesian Source Importance Estimation Approach for Multi-Source-Free Domain Adaptation (Han, AAAI 2023)
• Slides: date.pptx

• Presenter 1: Wang Jingge
• Paper: TR-GAN: Multi-Session Future MRI Prediction With Temporal Recurrent Generative Adversarial Network (Fan, IEEE Transactions on Medical Imaging 2022)
• Slides: tr-gan_multi-session_future_mri_prediction_with_temporal_recurrent_generative_adversarial_network.pptx

• Presenter 2: Zhao Zixi
• Paper: Max-Affine Spline Insights Into Deep Network Pruning (You and Balestriero, Transactions on Machine Learning Research 2022)
• Slides: 8.16_pre.pptx

• Presenter 1: Yang Jingyun
• Paper: Pick the Best Pre-trained Model: Towards Transferability Estimation for Medical Image Segmentation (Yang, MICCAI 2023)
• Slides: rd.pdf

• Presenter 2: Chen Xuechao
• Paper: Task-customized Masked Autoencoder via Mixture of Cluster-conditional Experts (Liu, ICLR 2023)
• Slides: paperreading20230830.pptx

• Presenter 1: Xiangyu Chen
• Paper: Graph Neural Networks can Recover the Hidden Features Solely from the Graph Structure (Sato, ICML 2023)
• Slides: graph_recover.pptx

• Presenter 2: Yanru Wu
• Paper: MATE: Plugging in Model Awareness to Task Embedding for Meta Learning (Chen and Wang, NeurIPS 2020)
• Slides: mate.pptx

• Presenter 1: Hanbing Liu
• Paper: Active Gradual Domain Adaptation: Dataset and Approach (Zhou, IEEE Transactions on Multimedia 2022)
• Slides: o

• Presenter 2: Jiahao Lai
• Paper: Hypergraph Neural Networks (Feng, AAAI 2019)
• Slides: hypergnn.pdf

• Presenter: Haohua Wang
• Paper: Fine-Tuning Language Models with Advantage-Induced Policy Alignment (Zhu, arXiv:2306.02231)
• Slides: fine-tuning_language_models_with_advantage-induced_policy_alignment_2_.pptx

—————————–