Histogram-based Parameter-efficient Tuning for Passive Sonar Classification

2025-04-21

AI Summary

In passive sonar classification, parameter-efficient transfer learning (PETL) methods struggle to model intermediate feature distribution shifts during domain adaptation. To address this, we propose a histogram-driven lightweight domain adaptation method. Our approach innovatively incorporates differentiable histogram statistics of input features into the tuning process, explicitly capturing target-domain distribution shifts—overcoming the insensitivity of conventional additive adapters to statistical drift. Integrated with a lightweight parameter modulation module, it is compatible with both Transformer and CNN backbones. Evaluated on the VTUAD dataset, our method achieves 91.8% classification accuracy—outperforming standard Adapter by 2.0%—while accelerating training and yielding feature representations closer to full fine-tuning. Crucially, it reduces trainable parameters by over 95%, enabling efficient, distribution-aware adaptation without compromising performance.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Parameter-efficient transfer learning (PETL) methods adapt large artificial neural networks to downstream tasks without fine-tuning the entire model. However, existing additive methods, such as adapters, sometimes struggle to capture distributional shifts in intermediate feature embeddings. We propose a novel histogram-based parameter-efficient tuning (HPT) technique that captures the statistics of the target domain and modulates the embeddings. Experimental results on three downstream passive sonar datasets (ShipsEar, DeepShip, VTUAD) demonstrate that HPT outperforms conventional adapters. Notably, HPT achieves 91.8% vs. 89.8% accuracy on VTUAD. Furthermore, HPT trains faster and yields feature representations closer to those of fully fine-tuned models. Overall, HPT balances parameter savings and performance, providing a distribution-aware alternative to existing adapters and shows a promising direction for scalable transfer learning in resource-constrained environments. The code is publicly available: https://github.com/Advanced-Vision-and-Learning-Lab/HLAST_DeepShip_ParameterEfficient.

Problem

• Improves parameter-efficient transfer learning for sonar classification
• Addresses distribution shifts in feature embeddings via histogram tuning
• Enhances accuracy and speed compared to conventional adapter methods

Innovation

• Histogram-based tuning for feature modulation
• Captures target domain statistics efficiently
• Outperforms adapters in accuracy and speed