Learning Unified Distance Metric Across Diverse Data Distributions with Parameter-Efficient Transfer Learning

• URL: http://arxiv.org/abs/2309.08944v2
• Date: Sun, 19 Jan 2025 03:07:21 GMT
• Title: Learning Unified Distance Metric Across Diverse Data Distributions with Parameter-Efficient Transfer Learning
• Authors: Sungyeon Kim, Donghyun Kim, Suha Kwak,
• Abstract summary: A common practice in metric learning is to train and test an embedding model for each dataset. This dataset-specific approach fails to simulate real-world scenarios that involve multiple heterogeneous distributions of data. We explore a new metric learning paradigm, called Unified Metric Learning (UML), which learns a unified distance metric.
• Score: 36.349282242221065
• License:
• Abstract: A common practice in metric learning is to train and test an embedding model for each dataset. This dataset-specific approach fails to simulate real-world scenarios that involve multiple heterogeneous distributions of data. In this regard, we explore a new metric learning paradigm, called Unified Metric Learning (UML), which learns a unified distance metric capable of capturing relations across multiple data distributions. UML presents new challenges, such as imbalanced data distribution and bias towards dominant distributions. These issues cause standard metric learning methods to fail in learning a unified metric. To address these challenges, we propose Parameter-efficient Unified Metric leArning (PUMA), which consists of a pre-trained frozen model and two additional modules, stochastic adapter and prompt pool. These modules enable to capture dataset-specific knowledge while avoiding bias towards dominant distributions. Additionally, we compile a new unified metric learning benchmark with a total of 8 different datasets. PUMA outperforms the state-of-the-art dataset-specific models while using about 69 times fewer trainable parameters.

Related papers

• Meta-Statistical Learning: Supervised Learning of Statistical Inference [59.463430294611626]
  This work demonstrates that the tools and principles driving the success of large language models (LLMs) can be repurposed to tackle distribution-level tasks. We propose meta-statistical learning, a framework inspired by multi-instance learning that reformulates statistical inference tasks as supervised learning problems.
  arXiv  Detail & Related papers  (2025-02-17T18:04:39Z)
• Multiply Robust Estimation for Local Distribution Shifts with Multiple Domains [9.429772474335122]
  We focus on scenarios where data distributions vary across multiple segments of the entire population. We propose a two-stage multiply robust estimation method to improve model performance on each individual segment. Our method is designed to be implemented with commonly used off-the-shelf machine learning models.
  arXiv  Detail & Related papers  (2024-02-21T22:01:10Z)
• Task-customized Masked AutoEncoder via Mixture of Cluster-conditional Experts [104.9871176044644]
  Masked Autoencoder(MAE) is a prevailing self-supervised learning method that achieves promising results in model pre-training. We propose a novel MAE-based pre-training paradigm, Mixture of Cluster-conditional Experts (MoCE) MoCE trains each expert only with semantically relevant images by using cluster-conditional gates.
  arXiv  Detail & Related papers  (2024-02-08T03:46:32Z)
• Federated Learning with Projected Trajectory Regularization [65.6266768678291]
  Federated learning enables joint training of machine learning models from distributed clients without sharing their local data. One key challenge in federated learning is to handle non-identically distributed data across the clients. We propose a novel federated learning framework with projected trajectory regularization (FedPTR) for tackling the data issue.
  arXiv  Detail & Related papers  (2023-12-22T02:12:08Z)
• FedSym: Unleashing the Power of Entropy for Benchmarking the Algorithms for Federated Learning [1.4656078321003647]
  Federated learning (FL) is a decentralized machine learning approach where independent learners process data privately. We study the currently popular data partitioning techniques and visualize their main disadvantages. We propose a method that leverages entropy and symmetry to construct 'the most challenging' and controllable data distributions.
  arXiv  Detail & Related papers  (2023-10-11T18:39:08Z)
• Learning from aggregated data with a maximum entropy model [73.63512438583375]
  We show how a new model, similar to a logistic regression, may be learned from aggregated data only by approximating the unobserved feature distribution with a maximum entropy hypothesis. We present empirical evidence on several public datasets that the model learned this way can achieve performances comparable to those of a logistic model trained with the full unaggregated data.
  arXiv  Detail & Related papers  (2022-10-05T09:17:27Z)
• Meta-Causal Feature Learning for Out-of-Distribution Generalization [71.38239243414091]
  This paper presents a balanced meta-causal learner (BMCL), which includes a balanced task generation module (BTG) and a meta-causal feature learning module (MCFL) BMCL effectively identifies the class-invariant visual regions for classification and may serve as a general framework to improve the performance of the state-of-the-art methods.
  arXiv  Detail & Related papers  (2022-08-22T09:07:02Z)
• A Personalized Federated Learning Algorithm: an Application in Anomaly Detection [0.6700873164609007]
  Federated Learning (FL) has recently emerged as a promising method to overcome data privacy and transmission issues. In FL, datasets collected from different devices or sensors are used to train local models (clients) each of which shares its learning with a centralized model (server) This paper proposes a novel Personalized FedAvg (PC-FedAvg) which aims to control weights communication and aggregation augmented with a tailored learning algorithm to personalize the resulting models at each client.
  arXiv  Detail & Related papers  (2021-11-04T04:57:11Z)
• Meta-Learned Confidence for Few-shot Learning [60.6086305523402]
  A popular transductive inference technique for few-shot metric-based approaches, is to update the prototype of each class with the mean of the most confident query examples. We propose to meta-learn the confidence for each query sample, to assign optimal weights to unlabeled queries. We validate our few-shot learning model with meta-learned confidence on four benchmark datasets.
  arXiv  Detail & Related papers  (2020-02-27T10:22:17Z)
• Learning Similarity Metrics for Numerical Simulations [29.39625644221578]
  We propose a neural network-based approach that computes a stable and generalizing metric (LSiM) to compare data from a variety of numerical simulation sources. Our method employs a Siamese network architecture that is motivated by the mathematical properties of a metric.
  arXiv  Detail & Related papers  (2020-02-18T20:11:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.