# M-ADDA: Unsupervised Domain Adaptation with Deep Metric Learning

arXiv: https://arxiv.org/abs/1807.02552 · HTML: https://mlsanity.com/paper/1807.02552/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Jul 2018.

## Built on by (most cited first, 8 of 8)

- [Deep Clustering: A Comprehensive Survey](https://mlsanity.com/paper/2210.04142/) (2022)
- [Instance Level Affinity-Based Transfer for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2104.01286/) (2021)
- [Attribute-Based Robotic Grasping with Data-Efficient Adaptation](https://mlsanity.com/paper/2501.02149/) (2025)
- [Multi-level Domain Adaptation for Lane Detection](https://mlsanity.com/paper/2206.10692/) (2022)
- [Learning Semantic Proxies from Visual Prompts for Parameter-Efficient Fine-Tuning in Deep Metric Learning](https://mlsanity.com/paper/2402.02340/) (2024)
- [Self-training via Metric Learning for Source-Free Domain Adaptation of Semantic Segmentation](https://mlsanity.com/paper/2212.04227/) (2022)
- [AdaTriplet-RA: Domain Matching via Adaptive Triplet and Reinforced Attention for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2211.08894/) (2022)
- [Weighted Unsupervised Domain Adaptation Considering Geometry Features and Engineering Performance of 3D Design Data](https://mlsanity.com/paper/2309.04499/) (2023)

## Abstract (arXiv)

Unsupervised domain adaptation techniques have been successful for a wide range of problems where supervised labels are limited. The task is to classify an unlabeled `target' dataset by leveraging a labeled `source' dataset that comes from a slightly similar distribution. We propose metric-based adversarial discriminative domain adaptation (M-ADDA) which performs two main steps. First, it uses a metric learning approach to train the source model on the source dataset by optimizing the triplet loss function. This results in clusters where embeddings of the same label are close to each other and those with different labels are far from one another. Next, it uses the adversarial approach (as that used in ADDA cite2017arXiv170205464T) to make the extracted features from the source and target datasets indistinguishable. Simultaneously, we optimize a novel loss function that encourages the target dataset's embeddings to form clusters. While ADDA and M-ADDA use similar architectures, we show that M-ADDA performs significantly better on the digits adaptation datasets of MNIST and USPS. This suggests that using metric-learning for domain adaptation can lead to large improvements in classification accuracy for the domain adaptation task. The code is available at urlhttps://github.com/IssamLaradji/M-ADDA.