Discriminatively boosted image clustering with fully convolutional auto-encoders

Li, Fengfu; Qiao, Hong; Zhang, Bo

doi:10.1016/j.patcog.2018.05.019

Cited by 223 publications

(120 citation statements)

References 18 publications

Supporting

Mentioning

120

Contrasting

Order By: Relevance

“…The first term can be viewed as a reconstruction term as it forces the inferred latent representation to recover its corresponding input and the second KL term can be considered as a regularization term to modulate the posterior of the learned representation to be Gaussian distribution. We used ReLu Instead of fully connected layers, a convolutional autoencoder (CAE) is equipped with convolutional layers in which each unit is connected to only local regions of the previous layer [22]. A convolutional layer consists of multiple filters (kernels) and each filter has a set of weights used to perform convolution operation that computes dot products between a filter and a local region [23].…”

Section: Deep Representation Learningmentioning

confidence: 99%

DeepMicro: deep representation learning for disease prediction based on microbiome data

Zhang

2019

Preprint

View full text Add to dashboard Cite

AbstractHuman microbiota plays a key role in human health and growing evidence supports the potential use of microbiome as a predictor of various diseases. However, the high-dimensionality of microbiome data, often in the order of hundreds of thousands, yet low sample sizes, poses great challenge for machine learning-based prediction algorithms. This imbalance induces the data to be highly sparse, preventing from learning a better prediction model. Also, there has been little work on deep learning applications to microbiome data with a rigorous evaluation scheme. To address these challenges, we propose DeepMicro, a deep representation learning framework allowing for an effective representation of microbiome profiles. DeepMicro successfully transforms high-dimensional microbiome data into a robust low-dimensional representation using various autoencoders and applies machine learning classification algorithms on the learned representation. In disease prediction, DeepMicro outperforms the current best approaches based on the strain-level marker profile in five different datasets. In addition, by significantly reducing the dimensionality of the marker profile, DeepMicro accelerates the model training and hyperparameter optimization procedure with 8X-30X speedup over the basic approach. DeepMicro is freely available at https://github.com/minoh0201/DeepMicro.

show abstract

Section: Deep Representation Learningmentioning

confidence: 99%

DeepMicro: deep representation learning for disease prediction based on microbiome data

Zhang

2019

Preprint

View full text Add to dashboard Cite

show abstract

“…Some proposals [25,56,60] jointly train an autoencoder neural network with a clustering algorithm, and use the internal representation provided by the autoencoder, i.e., the encoder output, as features for clustering. A different training method is used in [17,31,54], where autoencoders are initially pre-trained, and then fine-tuned using the cluster assignment loss. Finally, other techniques [24,57] combine clustering with standard convolutional neural networks (CNNs) for representation learning of images.…”

Section: Related Workmentioning

confidence: 99%

Differentially Private Mixture of Generative Neural Networks

Ács¹,

Melis²,

Castelluccia

et al. 2017

2017 IEEE International Conference on Data Mining (ICDM)

View full text Add to dashboard Cite

Generative models are used in a wide range of applications building on large amounts of contextually rich information. Due to possible privacy violations of the individuals whose data is used to train these models, however, publishing or sharing generative models is not always viable. In this paper, we present a novel technique for privately releasing generative models and entire high-dimensional datasets produced by these models. We model the generator distribution of the training data with a mixture of k generative neural networks. These are trained together and collectively learn the generator distribution of a dataset. Data is divided into k clusters, using a novel differentially private kernel k-means, then each cluster is given to separate generative neural networks, such as Restricted Boltzmann Machines or Variational Autoencoders, which are trained only on their own cluster using differentially private gradient descent. We evaluate our approach using the MNIST dataset, as well as call detail records and transit datasets, showing that it produces realistic synthetic samples, which can also be used to accurately compute arbitrary number of counting queries.

show abstract

“…[23] applies a variant of approximate K-means on features extracted with a pretrained AlexNet and [30] uses deep auto-encoders combined with ensemble clustering to generate feature representations suitable for clustering. As of today, a new family of methods consisting in learning jointly the clusters and the representation using alternating optimization has established a new baseline on the performance of image-set clustering algorithms ( [31,26,32]).…”

Section: Image-set Clusteringmentioning

confidence: 99%