paper-with-me

홈 › Papers

Information Flow in Self-Supervised Learning

2023-09-29 · Zhiquan Tan, Jingqin Yang, Weiran Huang, Yang Yuan, Yifan Zhang

In this paper, we conduct a comprehensive analysis of two dual-branch (Siamese architecture) self-supervised learning approaches, namely Barlow Twins and spectral contrastive learning, through the lens of matrix mutual information. We prove that the loss functions of these methods implicitly optimize both matrix mutual information and matrix joint entropy. This insight prompts us to further explore the category of single-branch algorithms, specifically MAE and U-MAE, for which mutual information and joint entropy become the entropy. Building on this intuition, we introduce the Matrix Variational Masked Auto-Encoder (M-MAE), a novel method that leverages the matrix-based estimation of entropy as a regularizer and subsumes U-MAE as a special case. The empirical evaluations underscore the effectiveness of M-MAE compared with the state-of-the-art methods, including a 3.9% improvement in linear probing ViT-Base, and a 1% improvement in fine-tuning ViT-Large, both on ImageNet.

📄 PDF Abstract BibTeX arXiv:2309.17281

Code (2)

yifanzhang-pro/m-mae 공식 구현 pytorch
huang-research-group/M-MAE pytorch

Tasks

Contrastive LearningSelf-Supervised Learning

Methods 이 논문이 사용한 방법론

Barlow Twins Barlow Twins is a self-supervised learning method that applies redundancy-reduction — a principle first proposed in neuroscience — to self supervised learning. The objective…
MAE 설명 없음

Similar Papers 제목 키워드 기반

SelFlow: Self-Supervised Learning of Optical Flow

2019-04-19 · CVPR 2019 6 · Pengpeng Liu, Michael Lyu, Irwin King, Jia Xu

We present a self-supervised learning approach for optical flow. Our method distills reliable flow estimations from non-occluded pixels, and uses these predictions as ground truth to learn optical flow for hallucinated o…

Optical Flow EstimationSelf-Supervised Learning

MC-JEPA: A Joint-Embedding Predictive Architecture for Self-Supervised Learning of Motion and Content Features

2023-07-24 · Adrien Bardes, Jean Ponce, Yann Lecun

Self-supervised learning of visual representations has been focusing on learning content features, which do not capture object motion or location, and focus on identifying and differentiating objects in images and videos…

Optical Flow EstimationSelf-Supervised LearningSemantic Segmentation

Self-supervised AutoFlow

2022-12-04 · CVPR 2023 1 · Hsin-Ping Huang, Charles Herrmann, Junhwa Hur, Erika Lu 외

Recently, AutoFlow has shown promising results on learning a training set for optical flow, but requires ground truth labels in the target domain to compute its search metric. Observing a strong correlation between the g…

Optical Flow Estimation

SeFlow: A Self-Supervised Scene Flow Method in Autonomous Driving

2024-07-01 · Qingwen Zhang, Yi Yang, Peizheng Li, Olov Andersson 외

Scene flow estimation predicts the 3D motion at each point in successive LiDAR scans. This detailed, point-level, information can help autonomous vehicles to accurately predict and understand dynamic changes in their sur…

Autonomous DrivingAutonomous VehiclesObjectScene Flow Estimation+1

Learning Scene Flow in 3D Point Clouds with Noisy Pseudo Labels

2022-03-23 · Bing Li, Cheng Zheng, Guohao Li, Bernard Ghanem

We propose a novel scene flow method that captures 3D motions from point clouds without relying on ground-truth scene flow annotations. Due to the irregularity and sparsity of point clouds, it is expensive and time-consu…

Pseudo LabelSelf-Supervised Learning