Study With Inha

ICLR 2023 Spotlight (notable-top-25%),(SparK) Designing BERT for Convolutional Networks: Sparse and Hierarchical Masked Modeling논문 링크: https://openreview.net/forum?id=NRxydtWup1SGitHub: https://github.com/keyu-tian/SparK Designing BERT for Convolutional Networks: Sparse and Hierarchical...This paper presents a simple yet powerful framework to pre-train convolutional network (convnet) with Sparse..

ICLR 2024 Oral Paper,Is ImageNet worth 1 video?Learning strong image encoders from 1 long unlabelled video논문 링크: https://openreview.net/forum?id=Yen1lGns2o Is ImageNet worth 1 video? Learning strong image encoders from 1...Self-supervised learning has unlocked the potential of scaling up pretraining to billions of images, since annotation is unnecessary. But are we making the best use of data? H..

Google Research, NeurIPS 2023 acceptedStableRep: Synthetic Images from Text-to-Image ModelsMake Strong Visual Representation Learners논문 링크: https://arxiv.org/pdf/2306.00984.pdf StableRep은 NeurIPS 2023에 accept된 논문으로 LG AI Research에서 정리한 NeurIPS 2023 주요 연구주제에 선정된 논문이다.LG AI 리서치 블로그: https://www.lgresearch.ai/blog/view?seq=379 [NeurIPS 2023] 주요 연구 주제와 주목할 만한 논문 소개 - LG AI Research BLOGNeurIPS 2023,..

CVPR 2020, Hyperbolic Image Embeddings 논문 링크: https://arxiv.org/abs/1904.02239(https://arxiv.org/abs/1904.02239) Hyperbolic Image Embeddings Computer vision tasks such as image classification, image retrieval and few-shot learning are currently dominated by Euclidean and spherical embeddings, so that the final decisions about class belongings or the degree of similarity are made using linear hy ..

Wu et al. / Federated Contrastive Learning for Volumetric Medical Image Segmentation / MICCAI 2021 Oral Federated Contrastive Learning for Volumetric Medical Image Segmentation 1. Problem Definition 해당 논문에서는 의료 영상으로 인공지능 모델을 학습할 때 겪는 대표적인 두 가지 문제를 제시했다. 레이블(label)이 있는 데이터로 학습을 시키는 지도 학습(Supervised Learning)은 많은 분야에서 좋은 결과를 보이고 있으나, 의료 데이터의 레이블을 구하기 위해서는 의료 전문가들이 필요하며 상당한 시간을 요구하기 때문에 레이블이 있는 방대한..

⚽ GOAL 2020 ~ 2023 사이에 활발하게 이루어진 연구들의 개념을 알아본다 각 개념의 대표적인 논문들을 간단하게 소개하여 연구의 흐름을 알아본다 이를 통해서 본인 연구/개발에서 써 볼만한 insight를 얻어갔으면 하는 마음.. 🙈 Unsupervised Learning : input data have no corresponding classifications or labeling examples Clustering (K-means…) Visualization and Dimensionality Reduction (PCA, t-SNE) 🙉 Semi-Supervised Learning : use a small set of input-output pairs and another set of only ..

Core-set: Active Learning for Convolutional Neural Networks 논문 링크: https://arxiv.org/abs/1708.00489 Active Learning for Convolutional Neural Networks: A Core-Set Approach Convolutional neural networks (CNNs) have been successfully applied to many recognition and learning tasks using a universal recipe; training a deep model on a very large dataset of supervised examples. However, this approach i..