Max-Margin Invariant Features from Transformed Unlabelled Data

2017-12-01 · NeurIPS 2017 12 · Dipan Pal, Ashwin Kannan, Gautam Arakalgud, Marios Savvides

The study of representations invariant to common transformations of the data is important to learning. Most techniques have focused on local approximate invariance implemented within expensive optimization frameworks lacking explicit theoretical guarantees. In this paper, we study kernels that are invariant to a unitary group while having theoretical guarantees in addressing the important practical issue of unavailability of transformed versions of labelled data. A problem we call the Unlabeled Transformation Problem which is a special form of semi-supervised learning and one-shot learning. We present a theoretically motivated alternate approach to the invariant kernel SVM based on which we propose Max-Margin Invariant Features (MMIF) to solve this problem. As an illustration, we design an framework for face recognition and demonstrate the efficacy of our approach on a large scale semi-synthetic dataset with 153,000 images and a new challenging protocol on Labelled Faces in the Wild (LFW) while out-performing strong baselines.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

Face Recognition One-Shot Learning

Methods 이 논문이 사용한 방법론

SVM A Support Vector Machine, or SVM, is a non-parametric supervised learning model. For non-linear classification and regression, they utilise the kernel trick to map inputs…

Max-Margin Invariant Features from Transformed Unlabeled Data

2017-10-24 · Dipan K. Pal, Ashwin A. Kannan, Gautam Arakalgud, Marios Savvides

Face Recognition One-Shot Learning

Neural View Synthesis and Matching for Semi-Supervised Few-Shot Learning of 3D Pose

2021-10-27 · NeurIPS 2021 12 · Angtian Wang, Shenxiao Mei, Alan Yuille, Adam Kortylewski

We study the problem of learning to estimate the 3D object pose from a few labelled examples and a collection of unlabelled data. Our main contribution is a learning framework, neural view synthesis and matching, that ca…

3D Pose Estimation Few-Shot Learning Pose Estimation

Learning invariant features using the Transformed Indian Buffet Process

2010-12-01 · NeurIPS 2010 12 · Joseph L. Austerweil, Thomas L. Griffiths

Identifying the features of objects becomes a challenge when those features can change in their appearance. We introduce the Transformed Indian Buffet Process (tIBP), and use it to define a nonparametric Bayesian model t…

Unsupervised Domain Adaptation for Brain Vessel Segmentation through Transwarp Contrastive Learning

2024-02-23 · Fengming Lin, Yan Xia, Michael MacRaild, Yash Deo 외

Unsupervised domain adaptation (UDA) aims to align the labelled source distribution with the unlabelled target distribution to obtain domain-invariant predictive models. Since cross-modality medical data exhibit signific…

Contrastive Learning Domain Adaptation Medical Image Analysis Unsupervised Domain Adaptation

Semi-Generative Modelling: Covariate-Shift Adaptation with Cause and Effect Features

2018-07-20 · Julius von Kügelgen, Alexander Mey, Marco Loog

Current methods for covariate-shift adaptation use unlabelled data to compute importance weights or domain-invariant features, while the final model is trained on labelled data only. Here, we consider a particular case o…

Domain Adaptation