A Unified Objective for Novel Class Discovery (ICLR2022) Robust Semi-Supervised Learning when Not All Classes have Labels (NIPS2022) OpenLDN: Learning to Discover Novel Classes for Open-World Semi-Supervised Learning (ECCV2022) Towards Realistic Semi-Supervised Learning(ECCV2022) 问题设定 输入:给定...
实验方面作者仅在MINST和CIFAR-10上做了验证,参考意义不大。 Open-World Semi-Supervised Learning(2020arXiv) === 这篇文章投ICLR2021没中,openreview链接: 作者在这篇文章中也提到了unlabeled set中可能出现novel class的情况,并且基于此场景提出了一种自动识别novel class的算法,取名为ORCA (stands for Open-woR...
这篇文章的思路其实在 PCHID rebuttal 的时候就基本完成了，投稿 2020 年初的 ICML 的时候某种意义上是糟了黑手。一月份完成的投稿，在 review 期间被问为什么没有 cite / 对比 Sergey Levine 组在 2019 年十二月底在 ICLR 上被 OpenReject 的工作（这篇就是后来的 GCSL：https://openreview.net/forum?id=By...
Trash To Treasure: Harvesting OOD Data With Cross-Modal Matching for Open-Set Semi-Supervised Learning. Junkai Huang, Chaowei Fang, Weikai Chen, Zhenhua Chai, Xiaolin Wei, Pengxu Wei, Liang Lin, Guanbin Li. (ICCV 2021) Energy-Based Open-World Uncertainty Modeling for Confidence Calibration. ...
『【ESPD】:Evolutionary Stochastic Policy Distillation.-增强版的PCHID;【SPLID】:ESPD的理论补全;ICLR’22:【WGCSL】』 ESPD 的工作流程 PCHID 过后一个自然的问题是:我们能不能替换掉这个 test function?能不能摆脱 Recursive Learning 的束缚?ESPD 给...
Multi-level Consistency Learning for Semi-supervised Domain Adaptation [IJCAI 2022] AdaMatch: A Unified Approach to Semi-Supervised Learning and Domain Adaptation [ICLR 2022] CLDA: Contrastive Learning for Semi-Supervised Domain Adaptation [NeurIPS] Deep Co-Training With Task Decomposition for Semi...
ICLR, CVPR, and ECCV, and Bioinformatics. The major keywords we use are “Biomedical Federated Learning, Medical Pretrained Foundation Model, Healthcare Federated Pretrain Training, etc”. The most representative papers like Med-BERT [24], FedClip [19], and MedClip [20] are regarded as seed...
[16] pioneers to consider open-world GAN attribution and discovery. However, their method follows a general semi-supervised learning pipeline, overlooking the intrinsic challenge of imperceptible trace discrimination. Our proposed POSE is tailored for this challenge and achie...
Angelova, Open-vocabulary object detection upon frozen vision and language models, in: ICLR, 2023. Google Scholar [66] Yao L., Han J., Wen Y., Liang X., Xu D., Zhang W., Li Z., Xu C., Xu H. Detclip: Dictionary-enriched visual-concept paralleled pre-training for open-world ...
发现PCHID背后的思想已经被人做了并且称为hindsight,但我用supervised learning搞RL仍然很novel的时候;在PCL的阳台上和丁哥打电话,厘清了PCHID的动态优化对应的时候;抱着随机过程的书,想明白state-space navigation 本质上就是缩短first hitting time的时候;为了快速迭代,设计的简单Maze环境在我尝试了所有SOTA算法数月...