Backward-compatible Aligned Representations Via An Orthogonal Transformation Layer

Abstract

Visual retrieval systems face significant challenges when updating models with improved representations due to misalignment between the old and new representations. The costly and resource-intensive backfilling process involves recalculating feature vectors for images in the gallery set whenever a new model is introduced. To address this, prior research has explored backward-compatible training methods that enable direct comparisons between new and old representations without backfilling. Despite these advancements, achieving a balance between backward compatibility and the performance of independently trained models remains an open problem. In this paper, we address it by expanding the representation space with additional dimensions and learning an orthogonal transformation to achieve compatibility with old models and, at the same time, integrate new information. This transformation preserves the original feature space's geometry, ensuring that our model aligns with previous versions

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyricci2024backward

Related papers

• Towards Backward-compatible Representation Learning (2020)12.93
• Forward Compatible Training For Large-scale Embedding Retrieval Systems (2021)8.09
• Boundary-aware Backward-compatible Representation Via Adversarial Learning In Image Retrieval (2023)9.21
• Prototype Perturbation For Relaxing Alignment Constraints In Backward-compatible Learning (2025)0.00
• Privacy-preserving Model Upgrades With Bidirectional Compatible Training In Image Retrieval (2022)0.00
• Metric Compatible Training For Online Backfilling In Large-scale Retrieval (2023)2.26
• Fastfill: Efficient Compatible Model Update (2023)0.00
• Learning Compatible Embeddings (2021)13.80