Computer Science
对比哈希
Contrastive HashingA machine learning technique that learns compact binary codes for data representation by maximizing the similarity between similar samples and minimizing it between dissimilar samples. / 一种通过最大化相似样本间相似性并最小化不相似样本间相似性来学习紧凑二进制代码的数据表示机器学习技术。
- Knowledge type
- 数学理论
- First proposed or discovered
- Proposer
- Turkey Farm member
- Creator
- 农场主
- Last updated
- 2026-11-14 23:06:37
Sub-knowledge points
Quantization LossSupervised LearningUnsupervised Learning
Higher-order knowledge points
Deep Learning / 工艺技术Representation Learning
Main parameters and performance indicators
Hashing Bits: The length of the binary code (e.g., 32, 64, 128).Contrastive Loss: A loss function that enforces the distance between positive pairs and negative pairs.
Open problems
Efficient training for large-scale datasets.Handling complex data distributions and noise.
Related literature
ranked by community starsCenter Contrastive Long-Tailed Hashing for Unsupervised Remote Sensing Image RetrievalOpen source
Related products and links
No item recorded yet
Creation and modification trail
1 recordsRecord title: Contrastive Hashing.
Node update feedback
review required