797,845 Hits in 3.8 sec

Learning to Transfer Learn: Reinforcement Learning-Based Selection for Adaptive Transfer Learning [article]

Linchao Zhu, Sercan O. Arik, Yi Yang, Tomas Pfister
2020 arXiv   pre-print
We propose a novel adaptive transfer learning framework, learning to transfer learn (L2TL), to improve performance on a target dataset by careful extraction of the related information from a source dataset  ...  We demonstrate that L2TL outperforms fine-tuning baselines and other adaptive transfer learning methods on eight datasets.  ...  Related Work Adaptive transfer learning: There is a long history of transfer learning for neural networks, particularly in the form of fine-tuning [12] .  ... 
arXiv:1908.11406v2 fatcat:ovmzwhgxzzckhl7jwyztskvjeq

Transfer Learning with Adaptive Regularizers [chapter]

Ulrich Rückert, Marius Kloft
2011 Lecture Notes in Computer Science  
We analytically investigate a moment-based method to obtain good values and give uniform convergence bounds for the prediction error on the target learning task.  ...  The success of regularized risk minimization approaches to classification with linear models depends crucially on the selection of a regularization term that matches with the learning task at hand.  ...  Regularization Adaptation with Transfer Learning Let us now describe the setting more formally. We are given a space of data objects X that are embedded in an Euclidean feature space, i.e.  ... 
doi:10.1007/978-3-642-23808-6_5 fatcat:nkpho2abhrfv5mkx4nnnggxmua

Concept Transfer Learning for Adaptive Language Understanding [article]

Su Zhu, Kai Yu
2019 arXiv   pre-print
Based on this new hierarchical representation, transfer learning approaches are developed for adaptive LU.  ...  To address this issue, in this paper, a novel concept transfer learning approach is proposed.  ...  We also present the concept transfer learning for adaptive LU on the atomic concepts level, to solve the problem of combinatory concepts extending in LU.  ... 
arXiv:1706.00927v3 fatcat:lnknun4jbfcddpzb2x6mymvscq

Transfer Adaptation Learning: A Decade Survey [article]

Lei Zhang, Xinbo Gao
2020 arXiv   pre-print
Broader solutions of transfer adaptation learning being created by researchers are identified, i.e., instance re-weighting adaptation, feature adaptation, classifier adaptation, deep network adaptation  ...  A research problem is characterized as transfer adaptation learning (TAL) when it needs knowledge correspondence between different moments/domains.  ...  ACKNOWLEDGMENT The author would like to thank the pioneer researchers in transfer learning, domain adaptation and other related fields. The author would also like to thank Dr. Mingsheng Long and Dr.  ... 
arXiv:1903.04687v2 fatcat:wurprqieffalnnp6isfkhh5y5i

Domain Adaptation Meets Disentangled Representation Learning and Style Transfer [article]

Hoang Tran Vu, Ching-Chun Huang
2018 arXiv   pre-print
The experimental results show that our architecture can be adaptive well to full transfer learning and partial transfer learning upon a well-learned disentangled representation.  ...  In this paper, a better learning network has been proposed by considering three tasks - domain adaptation, disentangled representation, and style transfer simultaneously.  ...  Unlike the conventional setting of domain adaptation, we pay further attention on (a) transfer learning of common feature components and (b) partial transfer learning.  ... 
arXiv:1712.09025v4 fatcat:w4japs2n2jdhno5k3p6ehjhsye

Hyperparameter Transfer Learning with Adaptive Complexity [article]

Samuel Horváth, Aaron Klein, Peter Richtárik, Cédric Archambeau
2021 arXiv   pre-print
In this work, we enable multi-task BO to compensate for this mismatch, such that the transfer learning procedure is able to handle different data regimes in a principled way.  ...  Bayesian optimization (BO) is a sample efficient approach to automatically tune the hyperparameters of machine learning models.  ...  We propose a two-step approach to do transfer learning in the context of Bayesian sequential decision making consisting in (1) an off-line learning of ordered shared representations and (2) an online adaptation  ... 
arXiv:2102.12810v1 fatcat:nlj6s6i4mvh6villxindirb24m

Adaptive Transfer Learning for Plant Phenotyping [article]

Jun Wu, Elizabeth A. Ainsworth, Sheng Wang, Kaiyu Guan, Jingrui He
2022 arXiv   pre-print
(2) Whether could the neural network based transfer learning models improve the performance of plant phenotyping?  ...  (3) Could the neural network based transfer learning be improved by using infinite-width hidden layers for plant phenotyping?  ...  To this end, we present a Gaussian process based adaptive transfer learning approach (Cao et al. 2010) for plant phenotyping.  ... 
arXiv:2201.05261v1 fatcat:soi2kbywlvfcxnie6w4ctgzf6u

Adapted tree boosting for Transfer Learning [article]

Wenjing Fang, Chaochao Chen, Bowen Song, Li Wang, Jun Zhou, Kenny Q. Zhu
2020 arXiv   pre-print
Inspired by this real case in Alipay, we view the problem as a transfer learning problem and design a set of revise strategies to transfer the source domain models to the target domain under the framework  ...  In this paper, we focus on the above mentioned realworld fraud dection problem and propose an adapted tree boosting workflow for transfer learning.  ...  Considering the above reasons, we choose XGBoost as the basic learner of this transfer learning framework and make adaptations to the source trees before continuing training.  ... 
arXiv:2002.11982v2 fatcat:l6s6dlsuzfaindewsfprzvxq54

Transfer Learning with Adaptive Fine-Tuning

Grega Vrbancic, Vili Podgorelec
2020 IEEE Access  
for the transfer learning approach being the multitask learning framework [35] .  ...  CONCLUSION In this work, we presented a novel DEFT adaptive method for transfer learning with fine-tuning, featuring the layer selection mechanism based on the DE algorithm.  ... 
doi:10.1109/access.2020.3034343 fatcat:dab2lvxmgvgkdhao6btbi4nfye

Balanced Distribution Adaptation for Transfer Learning [article]

Jindong Wang, Yiqiang Chen, Shuji Hao, Wenjie Feng, Zhiqi Shen
2018 arXiv   pre-print
To tackle the distribution adaptation problem, in this paper, we propose a novel transfer learning approach, named as Balanced Distribution Adaptation (BDA), which can adaptively leverage the importance  ...  Based on BDA, we also propose a novel Weighted Balanced Distribution Adaptation (W-BDA) algorithm to tackle the class imbalance issue in transfer learning.  ...  Balanced Distribution Adaptation Transfer learning methods often seek to adapt both the marginal and conditional distributions between domains [2], [7] .  ... 
arXiv:1807.00516v1 fatcat:ztl5i3dmjvfpvplccjoextgpqq

Transfer Learning and Deep Domain Adaptation [chapter]

Wen Xu, Jing He, Yanfeng Shu
2020 Advances and Applications in Deep Learning [Working Title]  
In particular deep domain adaptation (a branch of transfer learning) gets the most attention in recently published articles.  ...  Transfer learning is an emerging technique in machine learning, by which we can solve a new task with the knowledge obtained from an old task in order to address the lack of labeled data.  ...  Advances in Deep Learning Transfer Learning and Deep Domain Adaptation DOI: Transfer Learning and Deep Domain Adaptation DOI:  ... 
doi:10.5772/intechopen.94072 fatcat:xqqzz45hybg6lgue3bowqn6hke

Adaptive Feature Ranking for Unsupervised Transfer Learning [article]

Son N. Tran, Artur d'Avila Garcez
2014 arXiv   pre-print
Experiments carried out using the MNIST, ICDAR and TiCC image datasets show that the proposed adaptive feature ranking and transfer learning method offers statistically significant improvements on the  ...  Transfer Learning is concerned with the application of knowledge gained from solving a problem to a different but related problem domain.  ...  Feature Learning In this section, we introduce adaptive feature learning.  ... 
arXiv:1312.6190v2 fatcat:5obyuojcgbbhjnd5g4b2vfvb7m

An introduction to domain adaptation and transfer learning [article]

Wouter M. Kouw, Marco Loog
2019 arXiv   pre-print
Domain adaptation and transfer learning are sub-fields within machine learning that are concerned with accounting for these types of changes.  ...  We will start with a brief introduction into risk minimization, and how transfer learning and domain adaptation expand upon this framework.  ...  Such problem settings are known as domain adaptation or transfer learning settings [16, 160, 158] .  ... 
arXiv:1812.11806v2 fatcat:pkx3uhw4pbdwhcmzbvwxfvz2u4

Adaptive Transfer Learning on Graph Neural Networks [article]

Xueting Han, Zhenhuan Huang, Bang An, Jing Bai
2021 arXiv   pre-print
Conventional pre-training methods may be not effective enough on knowledge transfer since they do not make any adaptation for downstream tasks.  ...  Our methods can be applied to various transfer learning approaches, it performs well not only in multi-task learning but also in pre-training and fine-tuning.  ...  (i) We propose a new generic transfer learning paradigm on GNNs which adaptively selects and combines various auxiliary tasks on graphs.  ... 
arXiv:2107.08765v2 fatcat:ngnexsimgfezjfrvqhrtl4cslq

Deep Transfer Learning with Joint Adaptation Networks [article]

Mingsheng Long, Han Zhu, Jianmin Wang, Michael I. Jordan
2017 arXiv   pre-print
Deep networks have been successfully applied to learn transferable features for adapting models from a source domain to a different target domain.  ...  In this paper, we present joint adaptation networks (JAN), which learn a transfer network by aligning the joint distributions of multiple domain-specific layers across domains based on a joint maximum  ...  RTN jointly learns transferable features and adaptive classifiers by deep residual learning (He et al., 2016) .  ... 
arXiv:1605.06636v2 fatcat:gldukpal6nbrjeaikygqrizb2q
« Previous Showing results 1 — 15 out of 797,845 results