研究领域
- 暂无内容
论文成果 MORE+
- · Quantification and Analysis of Layer-wise and Pixel-wise Information Discarding.International Conference on Machine Learning.2022
- · RASAT: Integrating Relational Structures into Pretrained Seq2Seq Model for Text-to-SQL.Conference on Empirical Methods in Natural Language Processing.2022
- · RAPID DETECTION AND RECOGNITION OF WHOLE BRAIN ACTIVITY IN A FREELY BEHAVING Caenorhabditis elegans.PLOS Computational Biology.2022
- · Quantifying the Knowledge in a DNN to Explain Knowledge Distillation for Classification.IEEE Transactions on Pattern Analysis and Machine Intelligence.2022
- · Towards Theoretical Analysis of Transformation Complexity of ReLU DNNs.International Conference on Machine Learning.2022
- · Discovering and Explaining the Representation Bottleneck of DNNs.International Conference on Learning Representations.2022
- · Exploring Image Regions Not Well Encoded by an INN.INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 151.2022
- · Interpretable Generative Adversarial Networks.AAAI Conference on Artificial Intelligence.2021
- · A Unified Game-Theoretic Interpretation of Adversarial Robustness.Annual Conference on Neural Information Processing Systems.2021
- · Interpreting Attributions and Interactions of Adversarial Attacks.IEEE International Conferenceon Computer Vision.2021
专利
- 暂无内容
科研项目 MORE+
- · 深度模型表征诊断、解释、与定向修复的数学基础与若干证明-2026/12/01
- · 基于流形拟合的生成样本检测-2023/07/01
- · 基于博弈交互的语义概念解释与表达能力解释理论研究-2024/12/01
- · 人工智能前沿基础理论与关键技术-自主智能无人系统-机器智能-2023/12/01
- · 华为-上海交通大学智能多媒体技术长期合作框架协议-2021/11/01
- · 归因解释可视化算法的内在机理的解释与统一-2022/12/01
- · 博弈交互下深度视觉概念与特征鲁棒性的理论统一与解释-2024/06/01
- · 神经网络表达瓶颈与符号主义表征在博弈论下的统一与解释-2024/09/01
- · 国家级海外高层次人才引进计划项目-2024/01/01
- · AI可信属性的根因探索技术合作项目-2024/09/01