HashNet: Deep Learning to Hash by Continuation

Abstract

Learning to hash has been widely applied to approximate nearest neighbor search for large-scale multimedia retrieval, due to its computation efficiency and retrieval quality. Deep learning to hash, which improves retrieval quality by end-to-end representation learning and hash encoding, has received increasing attention recently. Subject to the ill-posed gradient difficulty in the optimization with sign activations, existing deep learning to hash methods need to first learn continuous representations and then generate binary hash codes in a separated binarization step, which suffer from substantial loss of retrieval quality. This work presents HashNet, a novel deep architecture for deep learning to hash by continuation method with convergence guarantees, which learns exactly binary hash codes from imbalanced similarity data. The key idea is to attack the ill-posed gradient problem in optimizing deep networks with non-smooth binary activations by continuation method, in which we begin from learning an easier network with smoothed activation function and let it evolve during the training, until it eventually goes back to being the original, difficult to optimize, deep network with the sign activation function. Comprehensive empirical evidence shows that HashNet can generate exactly binary hash codes and yield state-of-the-art multimedia retrieval performance on standard benchmarks.

References in corpus (3)

• Caffe: Convolutional Architecture for Fast Feature Embedding
• Supervised Discrete Hashing
• Deep Semantic Ranking Based Hashing for Multi-Label Image Retrieval

Cited by in corpus (18)

• Self-Supervised Video Hashing with Hierarchical Binary Auto-encoder
• Self-Binarizing Networks
• Candidate Generation with Binary Codes for Large-Scale Top-N Recommendation
• End-to-End Supervised Product Quantization for Image Search and Retrieval
• Deep Priority Hashing
• Learning Hash Codes via Hamming Distance Targets
• Hadamard Codebook Based Deep Hashing
• Climbing the WOL: Training for Cheaper Inference
• Push for Quantization: Deep Fisher Hashing
• SHREWD: Semantic Hierarchy-based Relational Embeddings for Weakly-supervised Deep Hashing
• Deep Hashing using Entropy Regularised Product Quantisation Network
• Deep Discrete Hashing with Self-supervised Pairwise Labels
• Deep Class-Wise Hashing: Semantics-Preserving Hashing via Class-wise Loss
• Recurrent Binary Embedding for GPU-Enabled Exhaustive Retrieval from Billion-Scale Semantic Vectors
• Using Deep Cross Modal Hashing and Error Correcting Codes for Improving the Efficiency of Attribute Guided Facial Image Retrieval
• Convolutional Hashing for Automated Scene Matching
• Hashing with Binary Matrix Pursuit
• Learning to hash with semantic similarity metrics and empirical KL divergence