Clustered Multi-Task Learning: A Convex Formulation.

We design a new spectral norm that encodes this a priori assumption, without the prior knowledge of the partition of tasks into groups, resulting in a new convex optimization formulation for multi-task ... We show in simulations on synthetic examples and on the IEDB MHC-I binding dataset, that our approach outperforms well-known convex methods for multi-task learning, as well as related non convex methods ... Conclusion We have presented a convex approach to clustered multi-task learning, based on the design of a dedicated norm. ...

arXiv:0809.2085v1 fatcat:3dv32fu3yzezza3j25vrsxzh4y

We consider the problem of multi-task learning in which tasks belong to hidden clusters. ... We formulate the learning problem as a novel convex optimization problem in which linear classifiers are combinations of (a small number of) some basis. ... Acknowledgments: This work was supported by a Marie Curie Reintegration Grant PIRG06-GA-2009-256497 and by the Intel Collaborative Research Institute for Computational Intelligence (ICRI-CI). ...

dblp:conf/aistats/BarzilaiC15 fatcat:l4amxa5danastfk6x2m7rs4tzi

Multi-task learning (MTL) aims to improve the performance of multiple related prediction tasks by leveraging useful information from them. ... Finally, we extend the proposed approach to a robust task clustering problem. ... Multi-task learning (MTL) is a common machine learning method that is used to solve this problem. ...

arXiv:2211.17204v1 fatcat:2tblwfcavnek3gdnzcbb6gun5q

This paper presents a multi-task learning approach by modeling the task-feature relationships. ... a co-cluster structure. ... We first formulate the problem of multi-task learning with task-feature co-clusters as a risk minimization problem with regularization to enforce the co-cluster structure of the task-feature relationships ...

doi:10.1609/aaai.v29i1.9483 fatcat:3et6vv57xfbrlissud3kkxukve

As an alternative MTL approach, clustered multi-task learning (CMTL) assumes that multiple tasks follow a clustered structure, i.e., tasks are partitioned into a set of groups where tasks in the same group ... Multi-task learning (MTL) learns multiple related tasks simultaneously to improve generalization performance. ... Multi-Task Learning: ASO and CMTL Assume we are given a multi-task learning problem with m tasks; each task i ∈ N m is associated with a set of training data {(x i 1 , y i 1 ), . . . , (x i ni , y i ni ...

pmid:25328366 pmcid:PMC4200604 fatcat:hz4c5x2mvrgk5gfjwbs33jvlbe

To address this issue, we propose a novel MTL method called Multi-Task Learning via Robust Regularized Clustering (MTLRRC). ... Multi-task learning (MTL) aims to improve estimation and prediction performance by sharing common information among related tasks. ... A. O. was supported by JST SPRING, Grant Number JPMJSP2136. S. K. was supported by JSPS KAKENHI Grant Numbers JP23K11008, JP23H03352, and JP23H00809. ...

arXiv:2404.03250v2 fatcat:cb2uamiojbchho2npowqzjopby

Multiple Versions

In this paper, we propose a multi-task clustering with model relation learning (MTCMRL) method, which automatically learns the model parameter relatedness between each pair of tasks. ... Multi-task clustering improves the clustering performance of each task by transferring knowledge among the related tasks. ... In light of the limitations of the existing multi-task clustering methods, in this paper, we propose a multi-task clustering with model relation learning (MTCMRL) method, which can automatically learn ...

doi:10.24963/ijcai.2018/435 dblp:conf/ijcai/ZhangZLL18 fatcat:z4pq7lpztff65flxyshkdw2umu

In this paper, we propose a regularization formulation for learning the relationships between tasks in multi-task learning. ... Multi-task learning is a learning paradigm which seeks to improve the generalization performance of a learning task with the help of some other related tasks. ... Our method can model global task relationships and the learning problem can be formulated directly as a convex optimization problem. ...

arXiv:1203.3536v1 fatcat:emvfr3qemvf5xjmxfxiqb6hkhq

In this paper, we propose a convex optimization formulation for learning from crowds without estimating the true labels by introducing personal models of the individual crowd workers. ... Recently crowdsourcing services are often used to collect a large amount of labeled data for machine learning, since they provide us an easy way to get labels at very low cost and in a short period. ... Multi-task learning is a learning task for simultaneously estimating multiple predictive models from multiple related tasks. The ideas of multi-task learning date back to the middle '90s. ...

doi:10.1609/aaai.v26i1.8105 fatcat:kbe4rtpsrjh2lokeq57bocho6y

The main contribution is a convex formulation that employs a graph-based regularizer and simultaneously discovers few groups of related tasks, having close-by task parameters, as well as the feature space ... Empirical results on benchmark datasets show that the proposed formulation achieves good generalization and outperforms state-of-the-art multi-task learning algorithms in some cases. ... Finds clusters of tasks having similar weight vectors. No feature learning is performed. 6 DMTL The multi-task feature learning formulation in Jalali et al. (2010) . ...

arXiv:1206.4611v1 fatcat:vczxybtakbgzbpwgwrvssblwmy

We present an improved formulation (iASO) for multi-task learning based on the non-convex alternating structure optimization (ASO) algorithm, in which all tasks are related by a shared feature representation ... Multi-task learning (MTL) aims to improve generalization performance by learning multiple related tasks simultaneously. ... A Convex Multi-Task Learning Formulation In this section, we consider a convex relaxation of the non-convex problem F 0 (iASO) in Eq. (3). ...

doi:10.1145/1553374.1553392 dblp:conf/icml/ChenTLY09 fatcat:2riqlfqbgfdsfj5yba3ghm3kaa

Multi-task clustering (MTC) has attracted a lot of research attentions in machine learning due to its ability in utilizing the relationship among different tasks. ... To alleviate these problems, we propose a novel self-paced multi-task clustering (SPMTC) paradigm. ... [24] proposed learning spectral kernel for multi-task clustering (LSKMTC) which learns a reproducing kernel hilbert space (RKHS) by formulating a unified kernel learning framework. ...

arXiv:1808.08068v1 fatcat:cl25e43imbctphkwbxy2tdnboq

To address this issue, we present Clustered Federated Learning (CFL), a novel Federated Multi-Task Learning (FMTL) framework, which exploits geometric properties of the FL loss surface, to group the client ... As clustering is only performed after Federated Learning has converged to a stationary point, CFL can be viewed as a post-processing method that will always achieve greater or equal performance than conventional ... CONCLUSION In this paper we presented Clustered Federated Learning, a framework for Federated Multi-Task Learning that can improve any existing Federated Learning Framework by enabling the participating ...

arXiv:1910.01991v1 fatcat:iabqswpkmfhwtmrshx33pn57li

In this paper, we propose a clustered multi-task learning, which can reveal and share the multi-task relationships for radar target recognition. ... classification [13], we propose a new classification method based on clustered multi-task learning theory. ... Clustered Multi-Task Learning (CMTL) [18] CMTL assumes that multiple tasks follow a clustered structure and that such a clustered structure is prior. ...

doi:10.3390/s17102218 pmid:28953267 pmcid:PMC5676668 fatcat:qq26hymeoncfzcvdhypr5f76ru

DOAJ

In this work, we present a novel and effective Multi-Task Information Bottleneck (MTIB) clustering method, which is capable of exploring the shared information between multiple action clustering tasks ... However, using shared information to improve multi-task human action clustering has never been considered before, and cannot be achieved using existing clustering methods. ... Zhang [33] proposes two convex multi-task clustering objectives, which aim to learn a shared feature representation and the task relationship, respectively. ...

doi:10.1109/cvpr.2017.431 dblp:conf/cvpr/YanHY17 fatcat:sp6yaej7w5bsxprun6n2al2zoq

Clustered Multi-Task Learning: A Convex Formulation [article]

Preserved Fulltext

Convex Multi-Task Learning by Clustering

Preserved Fulltext

Semisoft Task Clustering for Multi-Task Learning [article]

Preserved Fulltext

Exploiting Task-Feature Co-Clusters in Multi-Task Learning

Preserved Fulltext

Clustered Multi-Task Learning Via Alternating Structure Optimization

Preserved Fulltext

Multi-task learning via robust regularized clustering with non-convex group penalties [article]

Preserved Fulltext

Multi-Task Clustering with Model Relation Learning

Preserved Fulltext

A Convex Formulation for Learning Task Relationships in Multi-Task Learning [article]

Preserved Fulltext

A Convex Formulation for Learning from Crowds

Preserved Fulltext

A Convex Feature Learning Formulation for Latent Task Structure Discovery [article]

Preserved Fulltext

A convex formulation for learning shared structures from multiple tasks

Preserved Fulltext

Self-Paced Multi-Task Clustering [article]

Preserved Fulltext

Clustered Federated Learning: Model-Agnostic Distributed Multi-Task Optimization under Privacy Constraints [article]

Preserved Fulltext

Clustered Multi-Task Learning for Automatic Radar Target Recognition

Preserved Fulltext

Multi-task Clustering of Human Actions by Sharing Information

Preserved Fulltext