首页/论文/1906.01044

arXiv 1906.010442019-06cs.LGstat.ML

Weakly Supervised Disentanglement by Pairwise Similarities

University of Pittsburgh（匹兹堡大学）

Junxiang Chen, Kayhan Batmanghelich

英文摘要：

Recently, researches related to unsupervised disentanglement learning with deep generative models have gained substantial popularity. However, without introducing supervision, there is no guarantee that the factors of interest can be successfully recovered. Motivated by a real-world problem, we propose a setting where the user introduces weak supervision by providing similarities between instances based on a factor to be disentangled. The similarity is provided as either a binary (yes/no) or a real-valued label describing whether a pair of instances are similar or not. We propose a new method for weakly supervised disentanglement of latent variables within the framework of Variational Autoencoder. Experimental results demonstrate that utilizing weak supervision improves the performance of the disentanglement method substantially.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

阅读原文

阅读 PDF ↗

arXiv 页面 ↗HTML 原文 ↗

本文内容

英文摘要相关论文

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

继续探索

相关论文

Robust Ordinal VAE: Employing Noisy Pairwise Comparisons for Disentanglement

University of Pittsburgh（匹兹堡大学）

Recent work by Locatello et al. (2018) has shown that an inductive bias is required to disentangle factors of interest in Variational Autoencoder (VAE). Motivated by a real-world problem, we propose a setting where such bias is introduced by providing pairwise ordinal comparisons between instances, based on the desired factor to be disentangled. For example, a doctor compares pairs of patients based on the level of severity of their illnesses, and the desired factor is a quantitive level of the disease severity. In a real-world application, the pairwise comparisons are usually noisy. Our metho…

2019-10-14cs.LGstat.ML

3D Shape Variational Autoencoder Latent Disentanglement via Mini-Batch Feature Swapping for Bodies and Faces

University College London（伦敦大学学院）

Learning a disentangled, interpretable, and structured latent representation in 3D generative models of faces and bodies is still an open problem. The problem is particularly acute when control over identity features is required. In this paper, we propose an intuitive yet effective self-supervised approach to train a 3D shape variational autoencoder (VAE) which encourages a disentangled latent representation of identity features. Curating the mini-batch generation by swapping arbitrary features across different shapes allows to define a loss function leveraging known differences and similariti…

2022-03-23cs.CVcs.GRcs.LG

Classification from Pairwise Similarities/Dissimilarities and Unlabeled Data via Empirical Risk Minimization

The University of Tokyo（东京大学）；RIKEN（理化学研究所）

Pairwise similarities and dissimilarities between data points might be easier to obtain than fully labeled data in real-world classification problems, e.g., in privacy-aware situations. To handle such pairwise information, an empirical risk minimization approach has been proposed, giving an unbiased estimator of the classification risk that can be computed only from pairwise similarities and unlabeled data. However, this direction cannot handle pairwise dissimilarities so far. On the other hand, semi-supervised clustering is one of the methods which can use both similarities and dissimilaritie…

2019-04-26cs.LGstat.ML

基于成对相似度主动学习的相关聚类

Correlation Clustering with Active Learning of Pairwise Similarities

Chalmers University of Technology（查尔姆斯理工大学）

针对成对相似度未预先给出的相关聚类问题，提出通用的主动学习框架，具备反馈灵活、适应任意算法与查询策略及抗噪等优势，并设计了多种新颖查询策略，实验验证了其有效性。

2024-02-12cs.LGstat.ML

Hierarchical Clustering using Randomly Selected Similarities

Technicolor Research（特艺集团研究院）

The problem of hierarchical clustering items from pairwise similarities is found across various scientific disciplines, from biology to networking. Often, applications of clustering techniques are limited by the cost of obtaining similarities between pairs of items. While prior work has been developed to reconstruct clustering using a significantly reduced set of pairwise similarities via adaptive measurements, these techniques are only applicable when choice of similarities are available to the user. In this paper, we examine reconstructing hierarchical clustering under similarity observation…

2012-07-19cs.ITcs.LGmath.IT

Active Clustering: Robust and Efficient Hierarchical Clustering using Adaptively Selected Similarities

Boston University（波士顿大学）；University of Wisconsin - Madison（威斯康星大学麦迪逊分校）；Carnegie Mellon University（卡内基梅隆大学）

Hierarchical clustering based on pairwise similarities is a common tool used in a broad range of scientific applications. However, in many problems it may be expensive to obtain or compute similarities between the items to be clustered. This paper investigates the hierarchical clustering of N items based on a small subset of pairwise similarities, significantly less than the complete set of N(N-1)/2 similarities. First, we show that if the intracluster similarities exceed intercluster similarities, then it is possible to correctly determine the hierarchical clustering from as few as 3N log N s…

2011-02-18cs.ITcs.LGmath.IT

更多相关论文收起更多论文

使用成对相似性的自监督深度聚类

Deep Clustering with Self-Supervision using Pairwise Similarities

McGill University（麦吉尔大学）；Mila - Quebec AI Institute（米拉-魁北克人工智能研究所）

本文提出自监督深度聚类框架DCSS，先通过簇特定损失训练自编码器形成超球状潜在空间，再利用成对相似性构建K维空间以处理复杂簇分布，在七个基准数据集上验证了有效性。

2024-05-06cs.LG

用于数据聚类的成对相似度偏移

Shift of Pairwise Similarities for Data Clustering

Chalmers University of Technology（查尔姆斯理工大学）

本文提出通过对成对相似度进行自适应偏移来改进聚类，并设计高效局部搜索算法，实验表明其在多个数据集上优于常见聚类方法。

2025-02-05cs.AIcs.LG

Disentanglement of Correlated Factors via Hausdorff Factorized Support

University of Tübingen（图宾根大学）；ELLIS（欧洲学习与智能系统实验室）；Meta AI；FAIR（Meta基础人工智能研究院）；MPI for Intelligent Systems（马克斯·普朗克智能系统研究所）；MILA（蒙特利尔学习算法研究所）；Université de Montréal（蒙特利尔大学）；CIFAR（加拿大高等研究院）

A grand goal in deep learning research is to learn representations capable of generalizing across distribution shifts. Disentanglement is one promising direction aimed at aligning a model's representation with the underlying factors generating the data (e.g. color or background). Existing disentanglement methods, however, rely on an often unrealistic assumption: that factors are statistically independent. In reality, factors (like object color and shape) are correlated. To address this limitation, we consider the use of a relaxed disentanglement criterion -- the Hausdorff Factorized Support (H…

2023-02-25cs.LGstat.ML

连续直方图损失：超越神经相似性

Continuous Histogram Loss: Beyond Neural Similarity

ITMO University（ITMO大学）

本文提出 Continuous Histogram Loss (CHL) 损失函数，将二元相似性推广至连续多值相似性，通过可微二维直方图计算距离与相似性的条件概率，以解决相似性学习、表示学习和数据可视化等更广泛任务。

2020-04-06cs.LGstat.ML

Learning Latent Features with Pairwise Penalties in Low-Rank Matrix Completion

The Ohio State University（俄亥俄州立大学）；The University of Hong Kong（香港大学）；Carnegie Mellon University（卡内基梅隆大学）

Low-rank matrix completion has achieved great success in many real-world data applications. A matrix factorization model that learns latent features is usually employed and, to improve prediction performance, the similarities between latent variables can be exploited by pairwise learning using the graph regularized matrix factorization (GRMF) method. However, existing GRMF approaches often use the squared loss to measure the pairwise differences, which may be overly influenced by dissimilar pairs and lead to inferior prediction. To fully empower pairwise learning for matrix completion, we prop…

2020-01-26cs.LGstat.ML

冷启动主动相关聚类

Cold-Start Active Correlation Clustering

Chalmers University of Technology \& University of Gothenburg Sweden；Chalmers University of Technology \& University of Gothenburg

本文提出一种覆盖意识方法，解决冷启动场景下主动相关聚类中的成对相似性查询问题，通过合成和现实世界实验验证了方法的有效性。

2026-03-07cs.AIcs.LGcs.SI

Dimensionality Reduction via Diffusion Map Improved with Supervised Linear Projection

University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）

When performing classification tasks, raw high dimensional features often contain redundant information, and lead to increased computational complexity and overfitting. In this paper, we assume the data samples lie on a single underlying smooth manifold, and define intra-class and inter-class similarities using pairwise local kernel distances. We aim to find a linear projection to maximize the intra-class similarities and minimize the inter-class similarities simultaneously, so that the projected low dimensional data has optimized pairwise distances based on the label information, which is mor…

2020-08-08cs.CVcs.LGeess.IV

Embedding Transfer with Label Relaxation for Improved Metric Learning

POSTECH（浦项科技大学）

This paper presents a novel method for embedding transfer, a task of transferring knowledge of a learned embedding model to another. Our method exploits pairwise similarities between samples in the source embedding space as the knowledge, and transfers them through a loss used for learning target embedding models. To this end, we design a new loss called relaxed contrastive loss, which employs the pairwise similarities as relaxed labels for inter-sample relations. Our loss provides a rich supervisory signal beyond class equivalence, enables more important pairs to contribute more to training, …

2021-03-27cs.AIcs.CVcs.LG

C-AllOut: Catching & Calling Outliers by Type

University of São Paulo（圣保罗大学）；Carnegie Mellon University（卡内基梅隆大学）

Given an unlabeled dataset, wherein we have access only to pairwise similarities (or distances), how can we effectively (1) detect outliers, and (2) annotate/tag the outliers by type? Outlier detection has a large literature, yet we find a key gap in the field: to our knowledge, no existing work addresses the outlier annotation problem. Outliers are broadly classified into 3 types, representing distinct patterns that could be valuable to analysts: (a) global outliers are severe yet isolate cases that do not repeat, e.g., a data collection error; (b) local outliers diverge from their peers with…

2021-10-13cs.AIcs.LG

慢时变图中的低秩更新用于时空信号插值

Low-rank Updates in Slowly Time-varying Graphs for Spatial-Temporal Signal Interpolation

Department of EECS, York University（约克大学电子工程与计算机科学系）；Dept of ECE, University of Southern California（南加州大学电子工程系）

针对节点关系缓慢演变的时空数据，提出将相邻时刻邻接矩阵的差异建模为低秩矩阵，通过联合优化信号平滑先验和低秩先验进行信号插值，并利用快速正交匹配追踪算法近似低秩项，展开为轻量神经网络实现参数调优。

2026-06-24cs.LGeess.SP