scholarly journals Cross-modal Common Representation Learning by Hybrid Transfer Network

2017 ◽  
Author(s):  
Xin Huang ◽  
Yuxin Peng ◽  
Mingkuan Yuan
Keyword(s):  
Domain Knowledge ◽  
Large Scale ◽  
Representation Learning ◽  
Training Data ◽  
Target Domain ◽  
Retrieval Task ◽  
Source Domain ◽  
Useful Knowledge ◽  
Semantic Correlation ◽  
Common Representation

DNN-based cross-modal retrieval is a research hotspot to retrieve across different modalities as image and text, but existing methods often face the challenge of insufficient cross-modal training data. In single-modal scenario, similar problem is usually relieved by transferring knowledge from large-scale auxiliary datasets (as ImageNet). Knowledge from such single-modal datasets is also very useful for cross-modal retrieval, which can provide rich general semantic information that can be shared across different modalities. However, it is challenging to transfer useful knowledge from single-modal (as image) source domain to cross-modal (as image/text) target domain. Knowledge in source domain cannot be directly transferred to both two different modalities in target domain, and the inherent cross-modal correlation contained in target domain provides key hints for cross-modal retrieval which should be preserved during transfer process. This paper proposes Cross-modal Hybrid Transfer Network (CHTN) with two subnetworks: Modal-sharing transfer subnetwork utilizes the modality in both source and target domains as a bridge, for transferring knowledge to both two modalities simultaneously; Layer-sharing correlation subnetwork preserves the inherent cross-modal semantic correlation to further adapt to cross-modal retrieval task. Cross-modal data can be converted to common representation by CHTN for retrieval, and comprehensive experiment on 3 datasets shows its effectiveness.

scholarly journals CycleEmotionGAN: Emotional Semantic Consistency Preserved CycleGAN for Adapting Image Emotions

2019 ◽  
Vol 33 ◽  
pp. 2620-2627 ◽  
Author(s):  
Sicheng Zhao ◽  
Chuang Lin ◽  
Pengfei Xu ◽  
Sendong Zhao ◽  
Yuchen Guo ◽  
...  
Keyword(s):  
Large Scale ◽  
Domain Adaptation ◽  
Training Data ◽  
Emotion Classification ◽  
Target Domain ◽  
Adaptation Studies ◽  
Source Domain ◽  
Semantic Consistency ◽  
Annotation Information ◽  
Image Pairs

Deep neural networks excel at learning from large-scale labeled training data, but cannot well generalize the learned knowledge to new domains or datasets. Domain adaptation studies how to transfer models trained on one labeled source domain to another sparsely labeled or unlabeled target domain. In this paper, we investigate the unsupervised domain adaptation (UDA) problem in image emotion classification. Specifically, we develop a novel cycle-consistent adversarial model, termed CycleEmotionGAN, by enforcing emotional semantic consistency while adapting images cycleconsistently. By alternately optimizing the CycleGAN loss, the emotional semantic consistency loss, and the target classification loss, CycleEmotionGAN can adapt source domain images to have similar distributions to the target domain without using aligned image pairs. Simultaneously, the annotation information of the source images is preserved. Extensive experiments are conducted on the ArtPhoto and FI datasets, and the results demonstrate that CycleEmotionGAN significantly outperforms the state-of-the-art UDA approaches.

Knowledge Transfer with Weighted Adversarial Network for Cold-Start Store Site Recommendation

2021 ◽  
Vol 15 (3) ◽  
pp. 1-27
Author(s):  
Yan Liu ◽  
Bin Guo ◽  
Daqing Zhang ◽  
Djamal Zeghlache ◽  
Jingmin Chen ◽  
...  
Keyword(s):  
Large Scale ◽  
Cold Start ◽  
Weighting Scheme ◽  
Training Data ◽  
Chain Store ◽  
Useful Knowledge ◽  
Adversarial Network ◽  
Brick And Mortar ◽  
New City ◽  
Learning Machine

Store site recommendation aims to predict the value of the store at candidate locations and then recommend the optimal location to the company for placing a new brick-and-mortar store. Most existing studies focus on learning machine learning or deep learning models based on large-scale training data of existing chain stores in the same city. However, the expansion of chain enterprises in new cities suffers from data scarcity issues, and these models do not work in the new city where no chain store has been placed (i.e., cold-start problem). In this article, we propose a unified approach for cold-start store site recommendation, Weighted Adversarial Network with Transferability weighting scheme (WANT), to transfer knowledge learned from a data-rich source city to a target city with no labeled data. In particular, to promote positive transfer, we develop a discriminator to diminish distribution discrepancy between source city and target city with different data distributions, which plays the minimax game with the feature extractor to learn transferable representations across cities by adversarial learning. In addition, to further reduce the risk of negative transfer, we design a transferability weighting scheme to quantify the transferability of examples in source city and reweight the contribution of relevant source examples to transfer useful knowledge. We validate WANT using a real-world dataset, and experimental results demonstrate the effectiveness of our proposed model over several state-of-the-art baseline models.

scholarly journals X-MOL: large-scale pre-training for molecular understanding and diverse molecular analysis

2020 ◽  
Author(s):  
Dongyu Xue ◽  
Han Zhang ◽  
Dongling Xiao ◽  
Yukang Gong ◽  
Guohui Chuai ◽  
...  
Keyword(s):  
Molecular Analysis ◽  
In Silico ◽  
Large Scale ◽  
De Novo ◽  
Representation Learning ◽  
Training Data ◽  
Fine Tuning ◽  
Model Interpretation ◽  
Unlabelled Data ◽  
Super Computing

AbstractIn silico modelling and analysis of small molecules substantially accelerates the process of drug development. Representing and understanding molecules is the fundamental step for various in silico molecular analysis tasks. Traditionally, these molecular analysis tasks have been investigated individually and separately. In this study, we presented X-MOL, which applies large-scale pre-training technology on 1.1 billion molecules for molecular understanding and representation, and then, carefully designed fine-tuning was performed to accommodate diverse downstream molecular analysis tasks, including molecular property prediction, chemical reaction analysis, drug-drug interaction prediction, de novo generation of molecules and molecule optimization. As a result, X-MOL was proven to achieve state-of-the-art results on all these molecular analysis tasks with good model interpretation ability. Collectively, taking advantage of super large-scale pre-training data and super-computing power, our study practically demonstrated the utility of the idea of “mass makes miracles” in molecular representation learning and downstream in silico molecular analysis, indicating the great potential of using large-scale unlabelled data with carefully designed pre-training and fine-tuning strategies to unify existing molecular analysis tasks and substantially enhance the performance of each task.

scholarly journals Unsupervised Domain Adaptation by Statistics Alignment for Deep Sleep Staging Networks

2021 ◽  
Author(s):  
Jiahao Fan ◽  
Hangyu Zhu ◽  
Xinyu Jiang ◽  
Long Meng ◽  
Cong Fu ◽  
...  
Keyword(s):  
Large Scale ◽  
Domain Adaptation ◽  
Source Model ◽  
Deep Sleep ◽  
Sleep Staging ◽  
Target Domain ◽  
Source Domain ◽  
Unsupervised Domain Adaptation ◽  
Generalization Problem ◽  
Source Models

Deep sleep staging networks have reached top performance on large-scale datasets. However, these models perform poorer when training and testing on small sleep cohorts due to data inefficiency. Transferring well-trained models from large-scale datasets (source domain) to small sleep cohorts (target domain) is a promising solution but still remains challenging due to the domain-shift issue. In this work, an unsupervised domain adaptation approach, domain statistics alignment (DSA), is developed to bridge the gap between the data distribution of source and target domains. DSA adapts the source models on the target domain by modulating the domain-specific statistics of deep features stored in the Batch Normalization (BN) layers. Furthermore, we have extended DSA by introducing cross-domain statistics in each BN layer to perform DSA adaptively (AdaDSA). The proposed methods merely need the well-trained source model without access to the source data, which may be proprietary and inaccessible. DSA and AdaDSA are universally applicable to various deep sleep staging networks that have BN layers. We have validated the proposed methods by extensive experiments on two state-of-the-art deep sleep staging networks, DeepSleepNet+ and U-time. The performance was evaluated by conducting various transfer tasks on six sleep databases, including two large-scale databases, MASS and SHHS, as the source domain, four small sleep databases as the target domain. Thereinto, clinical sleep records acquired in Huashan Hospital, Shanghai, were used. The results show that both DSA and AdaDSA could significantly improve the performance of source models on target domains, providing novel insights into the domain generalization problem in sleep staging tasks.<br>

scholarly journals Learning supervised embeddings for large scale sequence comparisons

2019 ◽  
Author(s):  
Dhananjay Kimothi ◽  
Pravesh Biyani ◽  
James M Hogan ◽  
Akshay Soni ◽  
Wayne Kelly
Keyword(s):  
Large Scale ◽  
Representation Learning ◽  
Substantial Improvement ◽  
Target Space ◽  
Learning Sequence ◽  
Retrieval Task ◽  
Sequence Comparisons ◽  
Ongoing Research ◽  
Related Sequence ◽  
Hybrid Approaches

Similarity-based search of sequence collections is a core task in bioinformatics, one dominated for most of the genomic era by exact and heuristic alignment-based algorithms. However, even efficient heuristics such as BLAST may not scale to the data sets now emerging, motivating a range of alignment-free alternatives exploiting the underlying lexical structure of each sequence.In this paper, we introduce SuperVec, a novel supervised approach to learning sequence embeddings. Our method extends earlier Representation Learning (RL) based methods to include jointly contextual and class-related information for each sequence during training. This ensures that related sequence fragments have proximal representations in the target space, better reflecting the structure of the domain.Such representations may be used for downstream machine learning tasks or employed directly. Here, we apply SuperVec embeddings to a sequence retrieval task, where the goal is to retrieve sequences with the same family label as a given query. The SuperVec approach is extended further through H-SuperVec, a tree-based hierarchical method which learns embeddings across a range of feature spaces based on the class labels and their exclusive and exhaustive subsets.Experiments show that supervised learning of embeddings based on sequence labels using SuperVec and H-SuperVec provides a substantial improvement in retrieval performance over existing (unsupervised) RL-based approaches. Further, the new methods are an order of magnitude faster than BLAST for the database retrieval task, supporting hybrid approaches in which SuperVec rapidly filters the collection so that only potentially relevant records remain, allowing slower, more accurate methods to be executed quickly over a far smaller dataset. Thus, we may achieve faster query processing and higher precision than before.Finally, for some problems, direct use of embeddings is already sufficient to yield high levels of precision and recall. Extending this work to encompass weaker homology is the subject of ongoing research.

scholarly journals Distributional Correspondence Indexing for Cross-Lingual and Cross-Domain Sentiment Classification (Extended Abstract)

2018 ◽  
Author(s):  
Alejandro Moreo Fernández ◽  
Andrea Esuli ◽  
Fabrizio Sebastiani
Keyword(s):  
Domain Adaptation ◽  
State Of The Art ◽  
Sentiment Classification ◽  
Training Data ◽  
Target Domain ◽  
Source Domain ◽  
Machine Learning Methods ◽  
Cross Domain ◽  
Current State ◽  
Cross Lingual

Domain Adaptation (DA) techniques aim at enabling machine learning methods learn effective classifiers for a “target” domain when the only available training data belongs to a different “source” domain. In this extended abstract, we briefly describe our new DA method called Distributional Correspondence Indexing (DCI) for sentiment classification. DCI derives term representations in a vector space common to both domains where each dimension reflects its distributional correspondence to a pivot, i.e., to a highly predictive term that behaves similarly across domains. The experiments we have conducted show that DCI obtains better performance than current state-of-the-art techniques for cross-lingual and cross-domain sentiment classification.

scholarly journals Unsupervised Image-Generation Enhanced Adaptation for Object Detection in Thermal Images

2021 ◽  
Vol 2021 ◽  
pp. 1-6
Author(s):  
Peng Liu ◽  
Fuyu Li ◽  
Shanshan Yuan ◽  
Wanyi Li
Keyword(s):  
Object Detection ◽  
Unmanned Vehicles ◽  
Night Vision ◽  
Training Data ◽  
Image Generation ◽  
Target Domain ◽  
Image Domain ◽  
Thermal Images ◽  
Source Domain ◽  
Inversion Transformation

Object detection in thermal images is an important computer vision task and has many applications such as unmanned vehicles, robotics, surveillance, and night vision. Deep learning-based detectors have achieved major progress, which usually need large amount of labelled training data. However, labelled data for object detection in thermal images is scarce and expensive to collect. How to take advantage of the large number labelled visible images and adapt them into thermal image domain is expected to solve. This paper proposes an unsupervised image-generation enhanced adaptation method for object detection in thermal images. To reduce the gap between visible domain and thermal domain, the proposed method manages to generate simulated fake thermal images that are similar to the target images and preserves the annotation information of the visible source domain. The image generation includes a CycleGAN-based image-to-image translation and an intensity inversion transformation. Generated fake thermal images are used as renewed source domain, and then the off-the-shelf domain adaptive faster RCNN is utilized to reduce the gap between the generated intermediate domain and the thermal target domain. Experiments demonstrate the effectiveness and superiority of the proposed method.

scholarly journals ITERATIVE RE-WEIGHTED INSTANCE TRANSFER FOR DOMAIN ADAPTATION

2016 ◽  
Vol III-3 ◽  
pp. 339-346 ◽  
Author(s):  
A. Paul ◽  
F. Rottensteiner ◽  
C. Heipke
Keyword(s):  
Domain Adaptation ◽  
Training Data ◽  
Aerial Images ◽  
Target Domain ◽  
Data Set ◽  
Source Domain ◽  
The Difference ◽  
A New Technique ◽  
Class Labels ◽  
Decision Boundaries

Domain adaptation techniques in transfer learning try to reduce the amount of training data required for classification by adapting a classifier trained on samples from a source domain to a new data set (target domain) where the features may have different distributions. In this paper, we propose a new technique for domain adaptation based on logistic regression. Starting with a classifier trained on training data from the source domain, we iteratively include target domain samples for which class labels have been obtained from the current state of the classifier, while at the same time removing source domain samples. In each iteration the classifier is re-trained, so that the decision boundaries are slowly transferred to the distribution of the target features. To make the transfer procedure more robust we introduce weights as a function of distance from the decision boundary and a new way of regularisation. Our methodology is evaluated using a benchmark data set consisting of aerial images and digital surface models. The experimental results show that in the majority of cases our domain adaptation approach can lead to an improvement of the classification accuracy without additional training data, but also indicate remaining problems if the difference in the feature distributions becomes too large.

scholarly journals Autoencoder-based transfer learning in brain–computer interface for rehabilitation robot

2019 ◽  
Vol 16 (2) ◽  
pp. 172988141984086 ◽  
Author(s):  
Chuanqi Tan ◽  
Fuchun Sun ◽  
Bin Fang ◽  
Tao Kong ◽  
Wenchang Zhang
Keyword(s):  
Transfer Learning ◽  
Negative Transfer ◽  
Brain Computer Interface ◽  
Training Data ◽  
Computer Interface ◽  
Rehabilitation Robot ◽  
Target Domain ◽  
Source Domain ◽  
Adversarial Network ◽  
The Brain

The brain–computer interface-based rehabilitation robot has quickly become a very important research area due to its natural interaction. One of the most important problems in brain–computer interface is that large-scale annotated electroencephalography data sets required by advanced classifiers are almost impossible to acquire because biological data acquisition is challenging and quality annotation is costly. Transfer learning relaxes the hypothesis that the training data must be independent and identically distributed with the test data. It can be considered a powerful tool for solving the problem of insufficient training data. There are two basic issues with transfer learning, under transfer and negative transfer. We proposed a novel brain–computer interface framework by using autoencoder-based transfer learning, which includes three main components: an autoencoder framework, a joint adversarial network, and a regularized manifold constraint. The autoencoder framework automatically encodes and reconstructs data from source and target domains and forces the neural network to learn to represent these domains reliably. The joint adversarial network aims to force the network to learn to encode more appropriately for the source domain and target domain simultaneously, thereby overcoming the problem of under transfer. The regularized manifold constraint aims to avoid the problem of negative transfer by avoiding geometric manifold structure in the target domain being destroyed by the source domain. Experiments show that the brain–computer interface framework proposed by us can achieve better results than state-of-the-art approaches in electroencephalography signal classification tasks. This is helpful in aiding our rehabilitation robot to understand the intention of patients and can help patients to carry out rehabilitation exercises effectively.

scholarly journals Few shot domain adaptation for in situ macromolecule structural classification in cryoelectron tomograms

2020 ◽  
Author(s):  
Liangyong Yu ◽  
Ran Li ◽  
Xiangrui Zeng ◽  
Hongyi Wang ◽  
Jie Jin ◽  
...  
Keyword(s):  
Deep Learning ◽  
Large Scale ◽  
Spatial Organization ◽  
Domain Adaptation ◽  
Single Cells ◽  
Supplementary Information ◽  
Target Domain ◽  
Source Domain ◽  
Cellular Processes ◽  
Cross Domain

Abstract Motivation Cryoelectron tomography (cryo-ET) visualizes structure and spatial organization of macromolecules and their interactions with other subcellular components inside single cells in the close-to-native state at submolecular resolution. Such information is critical for the accurate understanding of cellular processes. However, subtomogram classification remains one of the major challenges for the systematic recognition and recovery of the macromolecule structures in cryo-ET because of imaging limits and data quantity. Recently, deep learning has significantly improved the throughput and accuracy of large-scale subtomogram classification. However, often it is difficult to get enough high-quality annotated subtomogram data for supervised training due to the enormous expense of labeling. To tackle this problem, it is beneficial to utilize another already annotated dataset to assist the training process. However, due to the discrepancy of image intensity distribution between source domain and target domain, the model trained on subtomograms in source domain may perform poorly in predicting subtomogram classes in the target domain. Results In this article, we adapt a few shot domain adaptation method for deep learning-based cross-domain subtomogram classification. The essential idea of our method consists of two parts: (i) take full advantage of the distribution of plentiful unlabeled target domain data, and (ii) exploit the correlation between the whole source domain dataset and few labeled target domain data. Experiments conducted on simulated and real datasets show that our method achieves significant improvement on cross domain subtomogram classification compared with baseline methods. Availability and implementation Software is available online https://github.com/xulabs/aitom. Supplementary information Supplementary data are available at Bioinformatics online.

Sign in / Sign up

Export Citation Format

Share Document