The Experts below are selected from a list of 56076 Experts worldwide ranked by ideXlab platform
Qiang Yang - One of the best experts on this subject based on the ideXlab platform.
-
PrivNet: Safeguarding Private Attributes in Transfer Learning for Recommendation
arXiv: Artificial Intelligence, 2020Co-Authors: Qiang YangAbstract:Transfer learning is an effective technique to improve a target recommender system with the knowledge from a Source Domain. Existing research focuses on the recommendation performance of the target Domain while ignores the privacy leakage of the Source Domain. The transferred knowledge, however, may unintendedly leak private information of the Source Domain. For example, an attacker can accurately infer user demographics from their historical purchase provided by a Source Domain Data owner. This paper addresses the above privacy-preserving issue by learning a privacy-aware neural representation by improving target performance while protecting Source privacy. The key idea is to simulate the attacks during the training for protecting unseen users' privacy in the future, modeled by an adversarial game, so that the transfer learning model becomes robust to attacks. Experiments show that the proposed PrivNet model can successfully disentangle the knowledge benefitting the transfer from leaking the privacy.
-
EMNLP (Findings) - PrivNet: Safeguarding Private Attributes in Transfer Learning for Recommendation
Findings of the Association for Computational Linguistics: EMNLP 2020, 2020Co-Authors: Qiang YangAbstract:Transfer learning is an effective technique to improve a target recommender system with the knowledge from a Source Domain. Existing research focuses on the recommendation performance of the target Domain while ignores the privacy leakage of the Source Domain. The transferred knowledge, however, may unintendedly leak private information of the Source Domain. For example, an attacker can accurately infer user demographics from their historical purchase provided by a Source Domain Data owner. This paper addresses the above privacy-preserving issue by learning a privacy-aware neural representation by improving target performance while protecting Source privacy. The key idea is to simulate the attacks during the training for protecting unseen users’ privacy in the future, modeled by an adversarial game, so that the transfer learning model becomes robust to attacks. Experiments show that the proposed PrivNet model can successfully disentangle the knowledge benefitting the transfer from leaking the privacy.
-
Selective Transfer Learning for Cross Domain Recommendation
Proceedings of the 2013 SIAM International Conference on Data Mining, 2013Co-Authors: Erheng Zhong, Lili Zhao, Weike Pan, Evan Wei Xiang, Qiang YangAbstract:Collaborative Filtering (CF) aims to predict users’ ratings on items according to historical user-itempreference Data. In many real-world applications, preference Data are usually sparse, which would make models overfit and fail to give accurate predictions. Recently, several research works show that by transferring knowledge from some manually selected Source Domains, the Data sparseness problem could be mitigated. However for most cases, parts of the Source Domain Data are not consistent with the observations in the target Domain, which may misguide the target Domain model building. In this paper, we propose a novel criterion based on empirical prediction error and its variance to capture the consistency across Domains in CF settings. Consequently, we embed this criterion into a boosting framework to perform selective knowledge transfer. Comparing with several state-of-the-art methods, we show that our proposed selective transfer learning framework can significantly improve the accuracy of rating prediction on several real-world recommendation tasks
-
SDM - Selective Transfer Learning for Cross Domain Recommendation
2013Co-Authors: Erheng Zhong, Lili Zhao, Weike Pan, Evan Wei Xiang, Qiang YangAbstract:Collaborative filtering (CF) aims to predict users’ ratings on items according to historical user-item preference Data. In many realworld applications, preference Data are usually sparse, which would make models overfit and fail to give accurate predictions. Recently, several research works show that by transferring knowledge from some manually selected Source Domains, the Data sparseness problem could be mitigated. However for most cases, parts of Source Domain Data are not consistent with the observations in the target Domain, which may misguide the target Domain model building. In this paper, we propose a novel criterion based on empirical prediction error and its variance to better capture the consistency across Domains in CF settings. Consequently, we embed this criterion into a boosting framework to perform selective knowledge transfer. Comparing to several state-of-the-art methods, we show that our proposed selective transfer learning framework can significantly improve the accuracy of rating prediction on several realworld recommendation tasks.
-
Selective Transfer Learning for Cross Domain Recommendation
arXiv: Learning, 2012Co-Authors: Erheng Zhong, Lili Zhao, Wei Xiang, Weike Pan, Qiang YangAbstract:Collaborative filtering (CF) aims to predict users' ratings on items according to historical user-item preference Data. In many real-world applications, preference Data are usually sparse, which would make models overfit and fail to give accurate predictions. Recently, several research works show that by transferring knowledge from some manually selected Source Domains, the Data sparseness problem could be mitigated. However for most cases, parts of Source Domain Data are not consistent with the observations in the target Domain, which may misguide the target Domain model building. In this paper, we propose a novel criterion based on empirical prediction error and its variance to better capture the consistency across Domains in CF settings. Consequently, we embed this criterion into a boosting framework to perform selective knowledge transfer. Comparing to several state-of-the-art methods, we show that our proposed selective transfer learning framework can significantly improve the accuracy of rating prediction tasks on several real-world recommendation tasks.
Gabriela Csurka - One of the best experts on this subject based on the ideXlab platform.
-
Domain Adaptation in Computer Vision Applications - What to Do When the Access to the Source Data Is Constrained
Domain Adaptation in Computer Vision Applications, 2017Co-Authors: Gabriela Csurka, Boris Chidlovskii, Stephane ClinchantAbstract:A large majority of existing Domain adaptation methods makes an assumption of freely available labeled Source and unlabeled target Data. They exploit the discrepancy between their distributions and build representations common to both target and Source Domains. In reality, such a simplifying assumption rarely holds, since Source Data are routinely a subject of legal and contractual constraints between Data owners and Data customers. Despite a limited access to Source Domain Data, decision-making procedures might be available in the form of, e.g., classification rules trained on the Source and made ready for a direct deployment and later reuse. In other cases, the owner of a Source Data is allowed to share a few representative examples such as class means. The aim of this chapter is therefore to address the Domain adaptation problem in such constrained real world applications, i.e. where the reuse of Source Domain Data is limited to classification rules or a few representative examples. As a solution, we extend recent techniques based on feature corruption and their marginalization, both considering supervised and unsupervised Domain adaptation settings. The proposed models are tested and compared on private and publicly available Source Datasets showing significant performance gains despite the absence of the whole Source Data and shortage of labeled target Data.
-
Domain adaptation in the absence of Source Domain Data
Knowledge Discovery and Data Mining, 2016Co-Authors: Boris Chidlovskii, Stephane Clinchant, Gabriela CsurkaAbstract:The overwhelming majority of existing Domain adaptation methods makes an assumption of freely available Source Domain Data. An equal access to both Source and target Data makes it possible to measure the discrepancy between their distributions and to build representations common to both target and Source Domains. In reality, such a simplifying assumption rarely holds, since Source Data are routinely a subject of legal and contractual constraints between Data owners and Data customers. When Source Domain Data can not be accessed, decision making procedures are often available for adaptation nevertheless. These procedures are often presented in the form of classification, identification, ranking etc. rules trained on Source Data and made ready for a direct deployment and later reuse. In other cases, the owner of a Source Data is allowed to share a few representative examples such as class means. In this paper we address the Domain adaptation problem in real world applications, where the reuse of Source Domain Data is limited to classification rules or a few representative examples. We extend the recent techniques of feature corruption and their marginalization, both in supervised and unsupervised settings. We test and compare them on private and publicly available Source Datasets and show that significant performance gains can be achieved despite the absence of Source Data and shortage of labeled target Data.
-
KDD - Domain Adaptation in the Absence of Source Domain Data
Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016Co-Authors: Boris Chidlovskii, Stephane Clinchant, Gabriela CsurkaAbstract:The overwhelming majority of existing Domain adaptation methods makes an assumption of freely available Source Domain Data. An equal access to both Source and target Data makes it possible to measure the discrepancy between their distributions and to build representations common to both target and Source Domains. In reality, such a simplifying assumption rarely holds, since Source Data are routinely a subject of legal and contractual constraints between Data owners and Data customers. When Source Domain Data can not be accessed, decision making procedures are often available for adaptation nevertheless. These procedures are often presented in the form of classification, identification, ranking etc. rules trained on Source Data and made ready for a direct deployment and later reuse. In other cases, the owner of a Source Data is allowed to share a few representative examples such as class means. In this paper we address the Domain adaptation problem in real world applications, where the reuse of Source Domain Data is limited to classification rules or a few representative examples. We extend the recent techniques of feature corruption and their marginalization, both in supervised and unsupervised settings. We test and compare them on private and publicly available Source Datasets and show that significant performance gains can be achieved despite the absence of Source Data and shortage of labeled target Data.
Boris Chidlovskii - One of the best experts on this subject based on the ideXlab platform.
-
Domain Adaptation in Computer Vision Applications - What to Do When the Access to the Source Data Is Constrained
Domain Adaptation in Computer Vision Applications, 2017Co-Authors: Gabriela Csurka, Boris Chidlovskii, Stephane ClinchantAbstract:A large majority of existing Domain adaptation methods makes an assumption of freely available labeled Source and unlabeled target Data. They exploit the discrepancy between their distributions and build representations common to both target and Source Domains. In reality, such a simplifying assumption rarely holds, since Source Data are routinely a subject of legal and contractual constraints between Data owners and Data customers. Despite a limited access to Source Domain Data, decision-making procedures might be available in the form of, e.g., classification rules trained on the Source and made ready for a direct deployment and later reuse. In other cases, the owner of a Source Data is allowed to share a few representative examples such as class means. The aim of this chapter is therefore to address the Domain adaptation problem in such constrained real world applications, i.e. where the reuse of Source Domain Data is limited to classification rules or a few representative examples. As a solution, we extend recent techniques based on feature corruption and their marginalization, both considering supervised and unsupervised Domain adaptation settings. The proposed models are tested and compared on private and publicly available Source Datasets showing significant performance gains despite the absence of the whole Source Data and shortage of labeled target Data.
-
Domain adaptation in the absence of Source Domain Data
Knowledge Discovery and Data Mining, 2016Co-Authors: Boris Chidlovskii, Stephane Clinchant, Gabriela CsurkaAbstract:The overwhelming majority of existing Domain adaptation methods makes an assumption of freely available Source Domain Data. An equal access to both Source and target Data makes it possible to measure the discrepancy between their distributions and to build representations common to both target and Source Domains. In reality, such a simplifying assumption rarely holds, since Source Data are routinely a subject of legal and contractual constraints between Data owners and Data customers. When Source Domain Data can not be accessed, decision making procedures are often available for adaptation nevertheless. These procedures are often presented in the form of classification, identification, ranking etc. rules trained on Source Data and made ready for a direct deployment and later reuse. In other cases, the owner of a Source Data is allowed to share a few representative examples such as class means. In this paper we address the Domain adaptation problem in real world applications, where the reuse of Source Domain Data is limited to classification rules or a few representative examples. We extend the recent techniques of feature corruption and their marginalization, both in supervised and unsupervised settings. We test and compare them on private and publicly available Source Datasets and show that significant performance gains can be achieved despite the absence of Source Data and shortage of labeled target Data.
-
KDD - Domain Adaptation in the Absence of Source Domain Data
Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016Co-Authors: Boris Chidlovskii, Stephane Clinchant, Gabriela CsurkaAbstract:The overwhelming majority of existing Domain adaptation methods makes an assumption of freely available Source Domain Data. An equal access to both Source and target Data makes it possible to measure the discrepancy between their distributions and to build representations common to both target and Source Domains. In reality, such a simplifying assumption rarely holds, since Source Data are routinely a subject of legal and contractual constraints between Data owners and Data customers. When Source Domain Data can not be accessed, decision making procedures are often available for adaptation nevertheless. These procedures are often presented in the form of classification, identification, ranking etc. rules trained on Source Data and made ready for a direct deployment and later reuse. In other cases, the owner of a Source Data is allowed to share a few representative examples such as class means. In this paper we address the Domain adaptation problem in real world applications, where the reuse of Source Domain Data is limited to classification rules or a few representative examples. We extend the recent techniques of feature corruption and their marginalization, both in supervised and unsupervised settings. We test and compare them on private and publicly available Source Datasets and show that significant performance gains can be achieved despite the absence of Source Data and shortage of labeled target Data.
Heidi Christensen - One of the best experts on this subject based on the ideXlab platform.
-
ICASSP - Source Domain Data Selection for Improved Transfer Learning Targeting Dysarthric Speech Recognition
ICASSP 2020 - 2020 IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP), 2020Co-Authors: Feifei Xiong, Jon Barker, Zhengjun Yue, Heidi ChristensenAbstract:This paper presents an improved transfer learning framework applied to robust personalised speech recognition models for speakers with dysarthria. As the baseline of transfer learning, a state-of-the-art CNN-TDNN-F ASR acoustic model trained solely on Source Domain Data is adapted onto the target Domain via neural network weight adaptation with the limited available Data from target dysarthric speakers. Results show that linear weights in neural layers play the most important role for an improved modelling of dysarthric speech evaluated using UASpeech corpus, achieving averaged 11.6% and 7.6% relative recognition improvement in comparison to the conventional speaker-dependent training and Data combination, respectively. To further improve the transferability towards target Domain, we propose an utterance-based Data selection of the Source Domain Data based on the entropy of posterior probability, which is analysed to statistically obey a Gaussian distribution. Compared to a speaker-based Data selection via dysarthria similarity measure, this allows for a more accurate selection of the potentially beneficial Source Domain Data for either increasing the target Domain training pool or constructing an intermediate Domain for incremental transfer learning, resulting in a further absolute recognition performance improvement of nearly 2% added to transfer learning baseline for speakers with moderate to severe dysarthria.
Patrick Perez - One of the best experts on this subject based on the ideXlab platform.
-
dada depth aware Domain adaptation in semantic segmentation
International Conference on Computer Vision, 2019Co-Authors: Himalaya Jain, Maxime Bucher, Matthieu Cord, Patrick PerezAbstract:Unsupervised Domain adaptation (UDA) is important for applications where large scale annotation of representative Data is challenging. For semantic segmentation in particular, it helps deploy on real “target Domain” Data models that are trained on annotated images from a different “Source Domain”, notably a virtual environment. To this end, most previous works consider semantic segmentation as the only mode of supervision for Source Domain Data, while ignoring other, possibly available, information like depth. In this work, we aim at exploiting at best such a privileged information while training the UDA model. We propose a unified depth-aware UDA framework that leverages in several complementary ways the knowledge of dense depth in the Source Domain. As a result, the performance of the trained semantic segmentation model on the target Domain is boosted. Our novel approach indeed achieves state-of-the-art performance on different challenging synthetic-2-real benchmarks.
-
ICCV - DADA: Depth-Aware Domain Adaptation in Semantic Segmentation
2019 IEEE CVF International Conference on Computer Vision (ICCV), 2019Co-Authors: Himalaya Jain, Maxime Bucher, Matthieu Cord, Patrick PerezAbstract:Unsupervised Domain adaptation (UDA) is important for applications where large scale annotation of representative Data is challenging. For semantic segmentation in particular, it helps deploy on real “target Domain” Data models that are trained on annotated images from a different “Source Domain”, notably a virtual environment. To this end, most previous works consider semantic segmentation as the only mode of supervision for Source Domain Data, while ignoring other, possibly available, information like depth. In this work, we aim at exploiting at best such a privileged information while training the UDA model. We propose a unified depth-aware UDA framework that leverages in several complementary ways the knowledge of dense depth in the Source Domain. As a result, the performance of the trained semantic segmentation model on the target Domain is boosted. Our novel approach indeed achieves state-of-the-art performance on different challenging synthetic-2-real benchmarks.