Shuffle And Learn: Minimizing Mutual Information For Unsupervised Hashing

Abstract

Unsupervised binary representation allows fast data retrieval without any annotations, enabling practical application like fast person re-identification and multimedia retrieval. It is argued that conflicts in binary space are one of the major barriers to high-performance unsupervised hashing as current methods failed to capture the precise code conflicts in the full domain. A novel relaxation method called Shuffle and Learn is proposed to tackle code conflicts in the unsupervised hash. Approximated derivatives for joint probability and the gradients for the binary layer are introduced to bridge the update from the hash to the input. Proof on \(\epsilon\)-Convergence of joint probability with approximated derivatives is provided to guarantee the preciseness on update applied on the mutual information. The proposed algorithm is carried out with iterative global updates to minimize mutual information, diverging the code before regular unsupervised optimization. Experiments suggest that t

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyliu2020shuffle

Related papers

• Multi-modal Mutual Information Maximization: A Novel Approach For Unsupervised Deep Cross-modal Hashing (2021)12.02
• Binary Representation Via Jointly Personalized Sparse Hashing (2022)9.59
• Supervised Hashing Based On Energy Minimization (2017)0.00
• Simultaneous Feature Aggregating And Hashing For Compact Binary Code Learning (2019)9.92
• Unsupervised Hashing With Contrastive Information Bottleneck (2021)13.50
• Simultaneous Compression And Quantization: A Joint Approach For Efficient Unsupervised Hashing (2018)7.50
• Collaborative Learning For Extremely Low Bit Asymmetric Hashing (2018)7.50
• Neurons Merging Layer: Towards Progressive Redundancy Reduction For Deep Supervised Hashing (2018)4.52