Large Margin Multi-Task Metric Learning

By Shibin Parameswaran, Kilian Q. Weinberger. 2010

296 citations recorded by the source corpus.

Subfields: Representation & Self-Supervised Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Representation & Self-Supervised Learning

Citations in the corpus

An Overview of Multi-task Learning (2018, methodology): “In earlier studies on this approach, task relations are either defined by model assumptions [36,37] or given by a priori information [38–41].”
A Survey on Multi-Task Learning (2017, methodology): “For example, multi-task metric learning [74], [113] transforms into an imbalanced binary classification problem where a pair of data points from the same class is treated as positive and that from different classes is negative.”
A Survey on Metric Learning for Feature Vectors and Structured Data (2013, methodology): “…learns a metric for each in a coupled fashion in order to improve the performance on all tasks. mt-LMNN (Parameswaran & Weinberger) Multi-Task LMNN24 (Parameswaran and Weinberger, 2010) is a straightforward adaptation of the ideas of Multi-Task SVM (Evgeniou and Pontil, 2004) to metric learning.”
Multi-Stage Multi-Task Feature Learning (2012, neutral): “It has been successfully applied to many applications such as speech classification (Parameswaran and Weinberger, 2010), handwritten character recognition (Obozinski et al., 2006; Quadrianto et al., 2010) and medical diagnosis (Bi et al., 2008).”
Interactive machine learning for health informatics: when do we need the human-in-the-loop? (2016, background): “This approach outperforms state-of-the-art MTL classiﬁers, however, many open research challenges remain open in this area [83].”
Learning Multi-Level Task Groups in Multi-Task Learning (2015, background): “MTL has gained its popularity among a wide range of applications including image annotation (Fan, Gao, and Luo 2008), speech recognition (Parameswaran and Weinberger 2010), disease progression predication (Zhou et al. 2011) and so on.”
Co-Training for Domain Adaptation (2011, methodology): “The final set of domain adaptation algorithms, which we compared against but did not describe, are those which actively seek to minimize the labeling divergence between domains using multi-task techniques [1, 8, 9, 12, 21, 27].”
Multi-Stage Multi-Task Learning with Reduced Rank (2016, background): “Luo 2008), speech recognition (Parameswaran and Weinberger 2010), age estimation (Zhang and Yeung 2010b), disease progression predication (Zhou et al.”
Heterogeneous-Neighborhood-based Multi-Task Local Learning Algorithms (2013, methodology): “…is denoted as MT-KNN, with the KNN classiﬁer which is a single-task learning method, the multi-task large margin nearest neighbor (mtLMNN) method [14] 1 which is a multi-task local learning method based on the homogeneous neighborhood, and the multi-task feature learning (MTFL) method [2] which…”
A Theoretical Analysis of Metric Hypothesis Transfer Learning (2015, background): “Following this idea, one of the most famous approach, called LMNN (Weinberger et al., 2005), proposes to learn a PSD matrix dedicated to improve the k-nearest neighbours algorithm.”
Distance-Based Image Classification: Generalizing to New Classes at Near-Zero Cost (2013, unknown): “In [38] various transfer learning methods were evaluated in a large-scale setting using the ILSVRC’10 dataset.”
Online Multitask Relative Similarity Learning (2017, unknown): “The most relevant one is [Parameswaran and Weinberger, 2010], in which a multi-task metric learning algorithm, namely mtLMNN, has been proposed by extending the popular LMNN algorithm [Weinberger and Saul, 2009] to multi-task learning scenarios. mtLMNN jointly learns a global metric for multiple…”
Multi-task deep visual-semantic embedding for video thumbnail selection (2015, methodology): “To solve the problem, we adopt the multi-task learning strategy, which refers to the joint training of multiple tasks, while enforcing a common intermediate parameterization or representation to improve each individual task’s performance [17].”
Learning with Whom to Share in Multi-task Feature Learning (2011, methodology): “The first one is to assume that the parameters used by all tasks are close to each other, either measured in the Frobenius norms of their differences (Evgeniou & Pontil, 2004; Liu et al., 2009; Zhang & Yeung, 2010; Parameswaran & Weinberger, 2010), or sharing a common prior (Yu et al., 2005; Lee et ”
Negative Margin Matters: Understanding Margin in Few-shot Classification (2020, background): “In the past decades, the idea of marginbased metric learning has been widely explored in SVM [40], k-NN classification [50], multi-task learning [33], etc.”
Multi-Agent Distributed Lifelong Learning for Collective Knowledge Acquisition (2017, background): “Existing methods in the literature have mostly investigated special cases of this setting for distributed multi-task learning (MTL) [7, 13, 25].”