Exploring Instance Relations for Unsupervised Feature Embedding
- URL: http://arxiv.org/abs/2105.03341v1
- Date: Fri, 7 May 2021 15:47:53 GMT
- Title: Exploring Instance Relations for Unsupervised Feature Embedding
- Authors: Yifei Zhang, Yu Zhou, Weiping Wang
- Abstract summary: In this paper, we explore instance relations including intra-instance multi-view relation and inter-instance relation for unsupervised feature embedding.
The proposed approach, referred to as EIR, is simple-yet-effective and can be easily inserted into existing view-invariant contrastive based methods.
- Score: 12.882929865091423
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Despite the great progress achieved in unsupervised feature embedding,
existing contrastive learning methods typically pursue view-invariant
representations through attracting positive sample pairs and repelling negative
sample pairs in the embedding space, while neglecting to systematically explore
instance relations. In this paper, we explore instance relations including
intra-instance multi-view relation and inter-instance interpolation relation
for unsupervised feature embedding. Specifically, we embed intra-instance
multi-view relation by aligning the distribution of the distance between an
instance's different augmented samples and negative samples. We explore
inter-instance interpolation relation by transferring the ratio of information
for image sample interpolation from pixel space to feature embedding space. The
proposed approach, referred to as EIR, is simple-yet-effective and can be
easily inserted into existing view-invariant contrastive learning based
methods. Experiments conducted on public benchmarks for image classification
and retrieval report state-of-the-art or comparable performance.
Related papers
- Deep Contrastive Multi-view Clustering under Semantic Feature Guidance [8.055452424643562]
We propose a multi-view clustering framework named Deep Contrastive Multi-view Clustering under Semantic feature guidance (DCMCS)
By minimizing instance-level contrastive loss weighted by semantic similarity, DCMCS adaptively weakens contrastive leaning between false negative pairs.
Experimental results on several public datasets demonstrate the proposed framework outperforms the state-of-the-art methods.
arXiv Detail & Related papers (2024-03-09T02:33:38Z) - Multi-level Relation Learning for Cross-domain Few-shot Hyperspectral
Image Classification [8.78907921615878]
Cross-domain few-shot hyperspectral image classification focuses on learning prior knowledge from a large number of labeled samples from source domains.
This paper proposes to learn sample relations on different levels and take them into the model learning process.
arXiv Detail & Related papers (2023-11-02T13:06:03Z) - Siamese Representation Learning for Unsupervised Relation Extraction [5.776369192706107]
Unsupervised relation extraction (URE) aims at discovering underlying relations between named entity pairs from open-domain plain text.
Existing URE models utilizing contrastive learning, which attract positive samples and repulse negative samples to promote better separation, have got decent effect.
We propose Siamese Representation Learning for Unsupervised Relation Extraction -- a novel framework to simply leverage positive pairs to representation learning.
arXiv Detail & Related papers (2023-10-01T02:57:43Z) - Soft Neighbors are Positive Supporters in Contrastive Visual
Representation Learning [35.53729744330751]
Contrastive learning methods train visual encoders by comparing views from one instance to others.
This binary instance discrimination is studied extensively to improve feature representations in self-supervised learning.
In this paper, we rethink the instance discrimination framework and find the binary instance labeling insufficient to measure correlations between different samples.
arXiv Detail & Related papers (2023-03-30T04:22:07Z) - Deep Incomplete Multi-view Clustering with Cross-view Partial Sample and
Prototype Alignment [50.82982601256481]
We propose a Cross-view Partial Sample and Prototype Alignment Network (CPSPAN) for Deep Incomplete Multi-view Clustering.
Unlike existing contrastive-based methods, we adopt pair-observed data alignment as 'proxy supervised signals' to guide instance-to-instance correspondence construction.
arXiv Detail & Related papers (2023-03-28T02:31:57Z) - Beyond Instance Discrimination: Relation-aware Contrastive
Self-supervised Learning [75.46664770669949]
We present relation-aware contrastive self-supervised learning (ReCo) to integrate instance relations.
Our ReCo consistently gains remarkable performance improvements.
arXiv Detail & Related papers (2022-11-02T03:25:28Z) - Deep Relational Metric Learning [84.95793654872399]
This paper presents a deep relational metric learning framework for image clustering and retrieval.
We learn an ensemble of features that characterizes an image from different aspects to model both interclass and intraclass distributions.
Experiments on the widely-used CUB-200-2011, Cars196, and Stanford Online Products datasets demonstrate that our framework improves existing deep metric learning methods and achieves very competitive results.
arXiv Detail & Related papers (2021-08-23T09:31:18Z) - Learning Multimodal VAEs through Mutual Supervision [72.77685889312889]
MEME combines information between modalities implicitly through mutual supervision.
We demonstrate that MEME outperforms baselines on standard metrics across both partial and complete observation schemes.
arXiv Detail & Related papers (2021-06-23T17:54:35Z) - An Unsupervised Sampling Approach for Image-Sentence Matching Using
Document-Level Structural Information [64.66785523187845]
We focus on the problem of unsupervised image-sentence matching.
Existing research explores to utilize document-level structural information to sample positive and negative instances for model training.
We propose a new sampling strategy to select additional intra-document image-sentence pairs as positive or negative samples.
arXiv Detail & Related papers (2021-03-21T05:43:29Z) - Out-of-distribution Generalization via Partial Feature Decorrelation [72.96261704851683]
We present a novel Partial Feature Decorrelation Learning (PFDL) algorithm, which jointly optimize a feature decomposition network and the target image classification model.
The experiments on real-world datasets demonstrate that our method can improve the backbone model's accuracy on OOD image classification datasets.
arXiv Detail & Related papers (2020-07-30T05:48:48Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.