Sparse-Dense Subspace Clustering

Shuai Yang, Wenqi Zhu, Yuesheng Zhu

Responsive image

Auto-TLDR; Sparse-Dense Subspace Clustering with Piecewise Correlation Estimation

Slides Poster

Subspace clustering refers to the problem of clustering high-dimensional data into a union of low-dimensional subspaces. Current subspace clustering approaches are usually based on a two-stage framework. In the first stage, an affinity matrix is generated from data. In the second one, spectral clustering is applied on the affinity matrix. However, the affinity matrix produced by two-stage methods cannot fully reveal the similarity between data points from the same subspace, resulting in inaccurate clustering. Besides, most approaches fail to solve large-scale clustering problems due to poor efficiency. In this paper, we first propose a new scalable sparse method called Iterative Maximum Correlation (IMC) to learn the affinity matrix from data. Then we develop Piecewise Correlation Estimation (PCE) to densify the intra-subspace similarity produced by IMC. Finally we extend our work into a Sparse-Dense Subspace Clustering (SDSC) framework with a dense stage to optimize the affinity matrix for two-stage methods. We show that IMC is efficient for large-scale tasks, and PCE ensures better performance for IMC. We show the universality of our SDSC framework for current two-stage methods as well. Experiments on benchmark data sets demonstrate the effectiveness of our approaches.

Similar papers

Fast Subspace Clustering Based on the Kronecker Product

Lei Zhou, Xiao Bai, Liang Zhang, Jun Zhou, Edwin Hancock

Responsive image

Auto-TLDR; Subspace Clustering with Kronecker Product for Large Scale Datasets

Slides Poster Similar

Subspace clustering is a useful technique for many computer vision applications in which the intrinsic dimension of high-dimensional data is often smaller than the ambient dimension. Spectral clustering, as one of the main approaches to subspace clustering, often takes on a sparse representation or a low-rank representation to learn a block diagonal self-representation matrix for subspace generation. However, existing methods require solving a large scale convex optimization problem with a large set of data, with computational complexity reaches O(N^3) for N data points. Therefore, the efficiency and scalability of traditional spectral clustering methods can not be guaranteed for large scale datasets. In this paper, we propose a subspace clustering model based on the Kronecker product. Due to the property that the Kronecker product of a block diagonal matrix with any other matrix is still a block diagonal matrix, we can efficiently learn the representation matrix which is formed by the Kronecker product of k smaller matrices. By doing so, our model significantly reduces the computational complexity to O(kN^{3/k}). Furthermore, our model is general in nature, and can be adapted to different regularization based subspace clustering methods. Experimental results on two public datasets show that our model significantly improves the efficiency compared with several state-of-the-art methods. Moreover, we have conducted experiments on synthetic data to verify the scalability of our model for large scale datasets.

Scalable Direction-Search-Based Approach to Subspace Clustering

Yicong He, George Atia

Responsive image

Auto-TLDR; Fast Direction-Search-Based Subspace Clustering

Slides Similar

Subspace clustering finds a multi-subspace representation that best fits a high-dimensional dataset. The computational and storage complexities of existing algorithms limit their usefulness for large scale data. In this paper, we develop a novel scalable approach to subspace clustering termed Fast Direction-Search-Based Subspace Clustering (Fast DiSC). In sharp contrast to existing scalable solutions which are mostly based on the self-expressiveness property of the data, Fast DiSC rests upon a new representation obtained from projections on computed data-dependent directions. These directions are derived from a convex formulation for optimal direction search to gauge hidden similarity relations. The computational complexity is significantly reduced by performing direction search in partitions of sampled data, followed by a retrieval step to cluster out-of-sample data using projections on the computed directions. A theoretical analysis underscores the ability of the proposed formulation to construct local similarity relations for the different data points. Experiments on both synthetic and real data demonstrate that the proposed algorithm can often outperform the state-of-the-art clustering methods.

Subspace Clustering Via Joint Unsupervised Feature Selection

Wenhua Dong, Xiaojun Wu, Hui Li, Zhenhua Feng, Josef Kittler

Responsive image

Auto-TLDR; Unsupervised Feature Selection for Subspace Clustering

Poster Similar

Any high-dimensional data arising from practical applications usually contains irrelevant features, which may impact on the performance of existing subspace clustering methods. This paper proposes a novel subspace clustering method, which reconstructs the feature matrix by the means of unsupervised feature selection (UFS) to achieve a better dictionary for subspace clustering (SC). Different from most existing clustering methods, the proposed approach uses a reconstructed feature matrix as the dictionary rather than the original data matrix. As the feature matrix reconstructed by representative features is more discriminative and closer to the ground-truth, it results in improved performance. The corresponding non-convex optimization problem is effectively solved using the half-quadratic and augmented Lagrange multiplier methods. Extensive experiments on four real datasets demonstrate the effectiveness of the proposed method.

Subspace Clustering for Action Recognition with Covariance Representations and Temporal Pruning

Giancarlo Paoletti, Jacopo Cavazza, Cigdem Beyan, Alessio Del Bue

Responsive image

Auto-TLDR; Unsupervised Learning for Human Action Recognition from Skeletal Data

Slides Similar

This paper tackles the problem of human action recognition, defined as classifying which action is displayed in a trimmed sequence, from skeletal data. Albeit state-of-the-art approaches designed for this application are all supervised, in this paper we pursue a more challenging direction: Solving the problem with unsupervised learning. To this end, we propose a novel subspace clustering method, which exploits covariance matrix to enhance the action’s discriminability and a timestamp pruning approach that allow us to better handle the temporal dimension of the data. Through a broad experimental validation, we show that our computational pipeline surpasses existing unsupervised approaches but also can result in favorable performances as compared to supervised methods.

A Spectral Clustering on Grassmann Manifold Via Double Low Rank Constraint

Xinglin Piao, Yongli Hu, Junbin Gao, Yanfeng Sun, Xin Yang, Baocai Yin

Responsive image

Auto-TLDR; Double Low Rank Representation for High-Dimensional Data Clustering on Grassmann Manifold

Slides Similar

High-dimension data clustering is a fundamental topic in machine learning and data mining areas. In recent year, researchers have proposed a series of effective methods based on Low Rank Representation (LRR) which could explore low-dimension subspace structure embedded in original data effectively. The traditional LRR methods usually treat original data as samples in Euclidean space. They generally adopt linear metric to measure the distance between two data. However, high-dimension data (such as video clip or imageset) are always considered as non-linear manifold data such as Grassmann manifold. Therefore, the traditional linear Euclidean metric would be no longer suitable for these special data. In addition, traditional LRR clustering method always adopt nuclear norm as low rank constraint which would lead to suboptimal solution and decrease the clustering accuracy. In this paper, we proposed a new low rank method on Grassmann manifold for high-dimension data clustering task. In the proposed method, a double low rank representation approach is proposed by combining the nuclear norm and bilinear representation for better construct the representation matrix. The experimental results on several public datasets show that the proposed method outperforms the state-of-the-art clustering methods.

Low Rank Representation on Product Grassmann Manifolds for Multi-viewSubspace Clustering

Jipeng Guo, Yanfeng Sun, Junbin Gao, Yongli Hu, Baocai Yin

Responsive image

Auto-TLDR; Low Rank Representation on Product Grassmann Manifold for Multi-View Data Clustering

Slides Poster Similar

Clustering high dimension multi-view data with complex intrinsic properties and nonlinear manifold structure is a challenging task since these data are always embedded in low dimension manifolds. Inspired by Low Rank Representation (LRR), some researchers extended classic LRR on Grassmann manifold or Product Grassmann manifold to represent data with non-linear metrics. However, most of these methods utilized convex nuclear norm to leverage a low-rank structure, which was over-relaxation of true rank and would lead to the results deviated from the true underlying ones. And, the computational complexity of singular value decomposition of matrix is high for nuclear norm minimization. In this paper, we propose a new low rank model for high-dimension multi-view data clustering on Product Grassmann Manifold with the matrix tri-factorization which is used to control the upper bound of true rank of representation matrix. And, the original problem can be transformed into the nuclear norm minimization with smaller scale matrices. An effective solution and theoretical analysis are also provided. The experimental results show that the proposed method obviously outperforms other state-of-the-art methods on several multi-source human/crowd action video datasets.

Constrained Spectral Clustering Network with Self-Training

Xinyue Liu, Shichong Yang, Linlin Zong

Responsive image

Auto-TLDR; Constrained Spectral Clustering Network: A Constrained Deep spectral clustering network

Slides Poster Similar

Deep spectral clustering networks have shown their superiorities due to the integration of feature learning and cluster assignment, and the ability to deal with non-convex clusters. Nevertheless, deep spectral clustering is still an ill-posed problem. Specifically, the affinity learned by the most remarkable SpectralNet is not guaranteed to be consistent with local invariance and thus hurts the final clustering performance. In this paper, we propose a novel framework of Constrained Spectral Clustering Network (CSCN) by incorporating pairwise constraints and clustering oriented fine-tuning to deal with the ill-posedness. To the best of our knowledge, this is the first constrained deep spectral clustering method. Another advantage of CSCN over existing constrained deep clustering networks is that it propagates pairwise constraints throughout the entire dataset. In addition, we design a clustering oriented loss by self-training to simultaneously finetune feature representations and perform cluster assignments, which further improve the quality of clustering. Extensive experiments on benchmark datasets demonstrate that our approach outperforms the state-of-the-art clustering methods.

Double Manifolds Regularized Non-Negative Matrix Factorization for Data Representation

Jipeng Guo, Shuai Yin, Yanfeng Sun, Yongli Hu

Responsive image

Auto-TLDR; Double Manifolds Regularized Non-negative Matrix Factorization for Clustering

Slides Poster Similar

Non-negative matrix factorization (NMF) is an important method in learning latent data representation. The local geometrical structure can make the learned representation more effectively and significantly improve the performance of NMF. However, most of existing graph-based learning methods are determined by a predefined similarity graph which may be not optimal for specific tasks. To solve the above the problem, we propose the Double Manifolds Regularized NMF (DMR-NMF) model which jointly learns an adaptive affinity matrix with the non-negative matrix factorization. The learned affinity matrix can guide the NMF to fit the clustering task. Moreover, we develop the iterative updating optimization schemes for DMR-NMF, and provide the strict convergence proof of our optimization strategy. Empirical experiments on four different real-world data sets demonstrate the state-of-the-art performance of DMR-NMF in comparison with the other related algorithms.

Label Self-Adaption Hashing for Image Retrieval

Jianglin Lu, Zhihui Lai, Hailing Wang, Jie Zhou

Responsive image

Auto-TLDR; Label Self-Adaption Hashing for Large-Scale Image Retrieval

Slides Poster Similar

Hashing has attracted widespread attention in image retrieval because of its fast retrieval speed and low storage cost. Compared with supervised methods, unsupervised hashing methods are more reasonable and suitable for large-scale image retrieval since it is always difficult and expensive to collect true labels of the massive data. Without label information, however, unsupervised hashing methods can not guarantee the quality of learned binary codes. To resolve this dilemma, this paper proposes a novel unsupervised hashing method called Label Self-Adaption Hashing (LSAH), which contains effective hashing function learning part and self-adaption label generation part. In the first part, we utilize anchor graph to keep the local structure of the data and introduce joint sparsity into the model to extract effective features for high-quality binary code learning. In the second part, a self-adaptive cluster label matrix is learned from the data under the assumption that the nearest neighbor points should have a large probability to be in the same cluster. Therefore, the proposed LSAH can make full use of the potential discriminative information of the data to guide the learning of binary code. It is worth noting that LSAH can learn effective binary codes, hashing function and cluster labels simultaneously in a unified optimization framework. To solve the resulting optimization problem, an Augmented Lagrange Multiplier based iterative algorithm is elaborately designed. Extensive experiments on three large-scale data sets indicate the promising performance of the proposed LSAH.

Graph Spectral Feature Learning for Mixed Data of Categorical and Numerical Type

Saswata Sahoo, Souradip Chakraborty

Responsive image

Auto-TLDR; Feature Learning in Mixed Type of Variable by an undirected graph

Slides Poster Similar

Feature learning in the presence of a mixed type of variables, numerical and categorical types, is important for related modeling problems. In this work, we propose a novel strategy to explicitly model the probabilistic dependence structure among the mixed type of variables by an undirected graph. The dependence structure among different pairs of variables are encoded by a suitable mapping function to estimate the edges of the graph. Spectral decomposition of the graph Laplacian provides the desired feature transformation. We numerically validate the implications of the feature learning strategy on various datasets in terms of data clustering.

Variational Deep Embedding Clustering by Augmented Mutual Information Maximization

Qiang Ji, Yanfeng Sun, Yongli Hu, Baocai Yin

Responsive image

Auto-TLDR; Clustering by Augmented Mutual Information maximization for Deep Embedding

Slides Poster Similar

Clustering is a crucial but challenging task in pattern analysis and machine learning. Recent many deep clustering methods combining representation learning with cluster techniques emerged. These deep clustering methods mainly focus on the correlation among samples and ignore the relationship between samples and their representations. In this paper, we propose a novel end-to-end clustering framework, namely variational deep embedding clustering by augmented mutual information maximization (VCAMI). From the perspective of VAE, we prove that minimizing reconstruction loss is equivalent to maximizing the mutual information of the input and its latent representation. This provides a theoretical guarantee for us to directly maximize the mutual information instead of minimizing reconstruction loss. Therefore we proposed the augmented mutual information which highlights the uniqueness of the representations while discovering invariant information among similar samples. Extensive experiments on several challenging image datasets show that the VCAMI achieves good performance. we achieve state-of-the-art results for clustering on MNIST (99.5%) and CIFAR-10 (65.4%) to the best of our knowledge.

Feature-Aware Unsupervised Learning with Joint Variational Attention and Automatic Clustering

Wang Ru, Lin Li, Peipei Wang, Liu Peiyu

Responsive image

Auto-TLDR; Deep Variational Attention Encoder-Decoder for Clustering

Slides Poster Similar

Deep clustering aims to cluster unlabeled real-world samples by mining deep feature representation. Most of existing methods remain challenging when handling high-dimensional data and simultaneously exploring the complementarity of deep feature representation and clustering. In this paper, we propose a novel Deep Variational Attention Encoder-decoder for Clustering (DVAEC). Our DVAEC improves the representation learning ability by fusing variational attention. Specifically, we design a feature-aware automatic clustering module to mitigate the unreliability of similarity calculation and guide network learning. Besides, to further boost the performance of deep clustering from a global perspective, we define a joint optimization objective to promote feature representation learning and automatic clustering synergistically. Extensive experimental results show the promising performance achieved by our DVAEC on six datasets comparing with several popular baseline clustering methods.

Learning Embeddings for Image Clustering: An Empirical Study of Triplet Loss Approaches

Kalun Ho, Janis Keuper, Franz-Josef Pfreundt, Margret Keuper

Responsive image

Auto-TLDR; Clustering Objectives for K-means and Correlation Clustering Using Triplet Loss

Slides Poster Similar

In this work, we evaluate two different image clustering objectives, k-means clustering and correlation clustering, in the context of Triplet Loss induced feature space embeddings. Specifically, we train a convolutional neural network to learn discriminative features by optimizing two popular versions of the Triplet Loss in order to study their clustering properties under the assumption of noisy labels. Additionally, we propose a new, simple Triplet Loss formulation, which shows desirable properties with respect to formal clustering objectives and outperforms the existing methods. We evaluate all three Triplet loss formulations for K-means and correlation clustering on the CIFAR-10 image classification dataset.

Motion Segmentation with Pairwise Matches and Unknown Number of Motions

Federica Arrigoni, Tomas Pajdla, Luca Magri

Responsive image

Auto-TLDR; Motion Segmentation using Multi-Modelfitting andpermutation synchronization

Slides Poster Similar

In this paper we address motion segmentation, that is the problem of clustering points in multiple images according to a number of moving objects. Two-frame correspondences are assumed as input without prior knowledge about trajectories. Our method is based on principles from ''multi-model fitting'' and ''permutation synchronization'', and - differently from previous techniques working under the same assumptions - it can handle an unknown number of motions. The proposed approach is validated on standard datasets, showing that it can correctly estimate the number of motions while maintaining comparable or better accuracy than the state of the art.

Sketch-Based Community Detection Via Representative Node Sampling

Mahlagha Sedghi, Andre Beckus, George Atia

Responsive image

Auto-TLDR; Sketch-based Clustering of Community Detection Using a Small Sketch

Slides Poster Similar

This paper proposes a sketch-based approach to the community detection problem which clusters the full graph through the use of an informative and concise sketch. The reduced sketch is built through an effective sampling approach which selects few nodes that best represent the complete graph and operates on a pairwise node similarity measure based on the average commute time. After sampling, the proposed algorithm clusters the nodes in the sketch, and then infers the cluster membership of the remaining nodes in the full graph based on their aggregate similarity to nodes in the partitioned sketch. By sampling nodes with strong representation power, our approach can improve the success rates over full graph clustering. In challenging cases with large node degree variation, our approach not only maintains competitive accuracy with full graph clustering despite using a small sketch, but also outperforms existing sampling methods. The use of a small sketch allows considerable storage savings, and computational and timing improvements for further analysis such as clustering and visualization. We provide numerical results on synthetic data based on the homogeneous, heterogeneous and degree corrected versions of the stochastic block model, as well as experimental results on real-world data.

Wasserstein k-Means with Sparse Simplex Projection

Takumi Fukunaga, Hiroyuki Kasai

Responsive image

Auto-TLDR; SSPW $k$-means: Sparse Simplex Projection-based Wasserstein $ k$-Means Algorithm

Slides Poster Similar

This paper presents a proposal of a faster Wasserstein $k$-means algorithm for histogram data by reducing Wasserstein distance computations exploiting sparse simplex projection. We shrink data samples, centroids and ground cost matrix, which enables significant reduction of the computations to solve optimal transport problems without loss of clustering quality. Furthermore, we dynamically reduce computational complexity by removing lower-valued data samples harnessing sparse simplex projection while keeping degradation of clustering quality lower. We designate this proposed algorithm as sparse simplex projection-based Wasserstein $k$-means, for short, SSPW $k$-means. Numerical evaluations against Wasserstein $k$-means algorithm demonstrate the effectiveness of the proposed SSPW $k$-means on real-world datasets.

Deep Superpixel Cut for Unsupervised Image Segmentation

Qinghong Lin, Weichan Zhong

Responsive image

Auto-TLDR; Deep Superpixel Cut for Deep Unsupervised Image Segmentation

Slides Poster Similar

Image segmentation, one of the most critical vision tasks, has been studied for many years. Most of the early algorithms are unsupervised methods, which use hand-crafted features to divide the image into many regions. Recently, owing to the great success of deep learning technology, CNNs based methods showing superior performance in image segmentation. However, these methods rely on a large number of human annotations, which are expensive to collect. In this paper, we propose a deep unsupervised method for image segmentation, which borrowed the ideas of classical graph partitioning. Our approach contains the following two stages. First, a Superpixel Guided Autoencoder (SGAE) is designed to learn the deep embedding and smooth the image simultaneously, then the smoothed image passed to generate superpixels. Second, based on the learned embedding, we propose a novel segmentation algorithm called Deep Superpixel Cut(DSC), which measures the deep similarity between superpixels and then adaptively partitions the superpixels into perceptual regions. Experimental results on the BSDS500 dataset demonstrate the effectiveness of the proposed method

JECL: Joint Embedding and Cluster Learning for Image-Text Pairs

Sean Yang, Kuan-Hao Huang, Bill Howe

Responsive image

Auto-TLDR; JECL: Clustering Image-Caption Pairs with Parallel Encoders and Regularized Clusters

Poster Similar

We propose JECL, a method for clustering image-caption pairs by training parallel encoders with regularized clustering and alignment objectives, simultaneously learning both representations and cluster assignments. These image-caption pairs arise frequently in high-value applications where structured training data is expensive to produce, but free-text descriptions are common. JECL trains by minimizing the Kullback-Leibler divergence between the distribution of the images and text to that of a combined joint target distribution and optimizing the Jensen-Shannon divergence between the soft cluster assignments of the images and text. Regularizers are also applied to JECL to prevent trivial solutions. Experiments show that JECL outperforms both single-view and multi-view methods on large benchmark image-caption datasets, and is remarkably robust to missing captions and varying data sizes.

A Multi-Task Multi-View Based Multi-Objective Clustering Algorithm

Sayantan Mitra, Sriparna Saha

Responsive image

Auto-TLDR; MTMV-MO: Multi-task multi-view multi-objective optimization for multi-task clustering

Slides Poster Similar

In recent years, multi-view multi-task clustering has received much attention. There are several real-life problems that involve both multi-view clustering and multi-task clustering, i.e., the tasks are closely related, and each task can be analyzed using multiple views. Traditional multi-task multi-view clustering algorithms use single-objective optimization-based approaches and cannot apply too-many regularization terms. However, these problems are inherently some multi-objective optimization problems because conflict may be between different views within a given task and also between different tasks, necessitating a trade-off. Based on these observations, in this paper, we have proposed a novel multi-task multi-view multi-objective optimization (MTMV-MO) algorithm which simultaneously optimizes three objectives, i.e., within-view task relation, within-task view relation and the quality of the clusters obtained. The proposed methodology (MTMV-MO) is evaluated on four different datasets and the results are compared with five state-of-the-art algorithms in terms of Adjusted Rand Index (ARI) and Classification Accuracy (%CoA). MTMV-MO illustrates an improvement of 1.5-2% in terms of ARI and 4-5% in terms of %CoA compared to the state-of-the-art algorithms.

Feature Extraction by Joint Robust Discriminant Analysis and Inter-Class Sparsity

Fadi Dornaika, Ahmad Khoder

Responsive image

Auto-TLDR; Robust Discriminant Analysis with Feature Selection and Inter-class Sparsity (RDA_FSIS)

Slides Similar

Feature extraction methods have been successfully applied to many real-world applications. The classical Linear Discriminant Analysis (LDA) and its variants are widely used as feature extraction methods. Although they have been used for different classification tasks, these methods have some shortcomings. The main one is that the projection axes obtained are not informative about the relevance of original features. In this paper, we propose a linear embedding method that merges two interesting properties: Robust LDA and inter-class sparsity. Furthermore, the targeted projection transformation focuses on the most discriminant original features. The proposed method is called Robust Discriminant Analysis with Feature Selection and Inter-class Sparsity (RDA_FSIS). Two kinds of sparsity are explicitly included in the proposed model. The first kind is obtained by imposing the $\ell_{2,1}$ constraint on the projection matrix in order to perform feature ranking. The second kind is obtained by imposing the inter-class sparsity constraint used for getting a common sparsity structure in each class. Comprehensive experiments on five real-world image datasets demonstrate the effectiveness and advantages of our framework over existing linear methods.

Cross-spectrum Face Recognition Using Subspace Projection Hashing

Hanrui Wang, Xingbo Dong, Jin Zhe, Jean-Luc Dugelay, Massimo Tistarelli

Responsive image

Auto-TLDR; Subspace Projection Hashing for Cross-Spectrum Face Recognition

Slides Poster Similar

Cross-spectrum face recognition, e.g. visible to thermal matching, remains a challenging task due to the large variation originated from different domains. This paper proposed a subspace projection hashing (SPH) to enable the cross-spectrum face recognition task. The intrinsic idea behind SPH is to project the features from different domains onto a common subspace, where matching the faces from different domains can be accomplished. Notably, we proposed a new loss function that can (i) preserve both inter-domain and intra-domain similarity; (ii) regularize a scaled-up pairwise distance between hashed codes, to optimize projection matrix. Three datasets, Wiki, EURECOM VIS-TH paired face and TDFace are adopted to evaluate the proposed SPH. The experimental results indicate that the proposed SPH outperforms the original linear subspace ranking hashing (LSRH) in the benchmark dataset (Wiki) and demonstrates a reasonably good performance for visible-thermal, visible-near-infrared face recognition, therefore suggests the feasibility and effectiveness of the proposed SPH.

Embedding Shared Low-Rank and Feature Correlation for Multi-View Data Analysis

Zhan Wang, Lizhi Wang, Hua Huang

Responsive image

Auto-TLDR; embedding shared low-rank and feature correlation for multi-view data analysis

Slides Poster Similar

The diversity of multimedia data in the real-world usually forms multi-view features. How to explore the structure information and correlations among multi-view features is still an open problem. In this paper, we propose a novel multi-view subspace learning method, named embedding shared low-rank and feature correlation (ESLRFC), for multi-view data analysis. First, in the embedding subspace, we propose a robust low-rank model on each feature set and enforce a shared low-rank constraint to characterize the common structure information of multiple feature data. Second, we develop an enhanced correlation analysis in the embedding subspace for simultaneously removing the redundancy of each feature set and exploring the correlations of multiple feature data. Finally, we incorporate the low-rank model and the correlation analysis into a unified framework. The shared low-rank constraint not only depicts the data distribution consistency among multiple feature data, but also assists robust subspace learning. Experimental results on recognition tasks demonstrate the superior performance and noise robustness of the proposed method.

Watermelon: A Novel Feature Selection Method Based on Bayes Error Rate Estimation and a New Interpretation of Feature Relevance and Redundancy

Xiang Xie, Wilhelm Stork

Responsive image

Auto-TLDR; Feature Selection Using Bayes Error Rate Estimation for Dynamic Feature Selection

Slides Poster Similar

Feature selection has become a crucial part of many classification problems in which high-dimensional datasets may contain tens of thousands of features. In this paper, we propose a novel feature selection method scoring the features through estimating the Bayes error rate based on kernel density estimation. Additionally, we update the scores of features dynamically by quantitatively interpreting the effects of feature relevance and redundancy in a new way. Distinguishing from the common heuristic applied by many feature selection methods, which prefers choosing features that are not relevant to each other, our approach penalizes only monotonically correlated features and rewards any other kind of relevance among features based on Spearman’s rank correlation coefficient and normalized mutual information. We conduct extensive experiments on seventeen diverse classification benchmarks, the results show that our approach overperforms other seventeen popular state-of-the-art feature selection methods in most cases.

Assortative-Constrained Stochastic Block Models

Daniel Gribel, Thibaut Vidal, Michel Gendreau

Responsive image

Auto-TLDR; Constrained Stochastic Block Models for Assortative Communities in Neural Networks

Slides Poster Similar

Stochastic block models (SBMs) are often used to find assortative community structures in networks, such that the probability of connections within communities is higher than in between communities. However, classic SBMs are not limited to assortative structures. In this study, we discuss the implications of this model-inherent indifference towards assortativity or disassortativity, and show that it can lead to undesirable outcomes in datasets which are known to be assortative but which contain a reduced amount of information. To circumvent these issues, we propose a constrained SBM that imposes strong assortativity constraints, along with efficient algorithmic solutions. These constraints significantly boost community-detection capabilities in regimes which are close to the detectability threshold. They also permit to identify structurally-different communities in networks representing cerebral-cortex activity regions.

Classification and Feature Selection Using a Primal-Dual Method and Projections on Structured Constraints

Michel Barlaud, Antonin Chambolle, Jean_Baptiste Caillau

Responsive image

Auto-TLDR; A Constrained Primal-dual Method for Structured Feature Selection on High Dimensional Data

Slides Poster Similar

This paper deals with feature selection using supervised classification on high dimensional datasets. A classical approach is to project data on a low dimensional space and classify by minimizing an appropriate quadratic cost. Our first contribution is to introduce a matrix of centers in the definition of this cost. Moreover, as quadratic costs are not robust to outliers, we propose to use an $\ell_1$ cost instead (or Huber loss to mitigate overfitting issues). While control on sparsity is commonly obtained by adding an $\ell_1$ constraint on the vectorized matrix of weights used for projecting the data, our second contribution is to enforce structured sparsity. To this end we propose constraints that take into account the matrix structure of the data, based either on the nuclear norm, on the $\ell_{2,1}$ norm, or on the $\ell_{1,2}$ norm for which we provide a new projection algorithm. We optimize simultaneously the projection matrix and the matrix of centers thanks to a new tailored constrained primal-dual method. The primal-dual framework is general enough to encompass the various robust losses and structured constraints we use, and allows a convergence analysis. We demonstrate the effectiveness of the approach on three biological datasets. Our primal-dual method with robust losses, adaptive centers and structured constraints does significantly better than classical methods, both in terms of accuracy and computational time.

N2D: (Not Too) Deep Clustering Via Clustering the Local Manifold of an Autoencoded Embedding

Ryan Mcconville, Raul Santos-Rodriguez, Robert Piechocki, Ian Craddock

Responsive image

Auto-TLDR; Local Manifold Learning for Deep Clustering on Autoencoded Embeddings

Slides Similar

Deep clustering has increasingly been demonstrating superiority over conventional shallow clustering algorithms. Deep clustering algorithms usually combine representation learning with deep neural networks to achieve this performance, typically optimizing a clustering and non-clustering loss. In such cases, an autoencoder is typically connected with a clustering network, and the final clustering is jointly learned by both the autoencoder and clustering network. Instead, we propose to learn an autoencoded embedding and then search this further for the underlying manifold. For simplicity, we then cluster this with a shallow clustering algorithm, rather than a deeper network. We study a number of local and global manifold learning methods on both the raw data and autoencoded embedding, concluding that UMAP in our framework is able to find the best clusterable manifold of the embedding. This suggests that local manifold learning on an autoencoded embedding is effective for discovering higher quality clusters. We quantitatively show across a range of image and time-series datasets that our method has competitive performance against the latest deep clustering algorithms, including out-performing current state-of-the-art on several. We postulate that these results show a promising research direction for deep clustering. The code can be found at https://github.com/rymc/n2d.

Soft Label and Discriminant Embedding Estimation for Semi-Supervised Classification

Fadi Dornaika, Abdullah Baradaaji, Youssof El Traboulsi

Responsive image

Auto-TLDR; Semi-supervised Semi-Supervised Learning for Linear Feature Extraction and Label Propagation

Slides Poster Similar

In recent times, graph-based semi-supervised learning proved to be a powerful paradigm for processing and mining large datasets. The main advantage relies on the fact that these methods can be useful in propagating a small set of known labels to a large set of unlabeled data. The scarcity of labeled data may affect the performance of the semi-learning. This paper introduces a new semi-supervised framework for simultaneous linear feature extraction and label propagation. The proposed method simultaneously estimates a discriminant transformation and the unknown label by exploiting both labeled and unlabeled data. In addition, the unknowns of the learning model are estimated by integrating two types of graph-based smoothness constraints. The resulting semi-supervised model is expected to learn more discriminative information. Experiments are conducted on six public image datasets. These experimental results show that the performance of the proposed method can be better than that of many state-of-the-art graph-based semi-supervised algorithms.

Local Clustering with Mean Teacher for Semi-Supervised Learning

Zexi Chen, Benjamin Dutton, Bharathkumar Ramachandra, Tianfu Wu, Ranga Raju Vatsavai

Responsive image

Auto-TLDR; Local Clustering for Semi-supervised Learning

Slides Similar

The Mean Teacher (MT) model of Tarvainen and Valpola has shown favorable performance on several semi-supervised benchmark datasets. MT maintains a teacher model's weights as the exponential moving average of a student model's weights and minimizes the divergence between their probability predictions under diverse perturbations of the inputs. However, MT is known to suffer from confirmation bias, that is, reinforcing incorrect teacher model predictions. In this work, we propose a simple yet effective method called Local Clustering (LC) to mitigate the effect of confirmation bias. In MT, each data point is considered independent of other points during training; however, data points are likely to be close to each other in feature space if they share similar features. Motivated by this, we cluster data points locally by minimizing the pairwise distance between neighboring data points in feature space. Combined with a standard classification cross-entropy objective on labeled data points, the misclassified unlabeled data points are pulled towards high-density regions of their correct class with the help of their neighbors, thus improving model performance. We demonstrate on semi-supervised benchmark datasets SVHN and CIFAR-10 that adding our LC loss to MT yields significant improvements compared to MT and performance comparable to the state of the art in semi-supervised learning.

Aggregating Dependent Gaussian Experts in Local Approximation

Hamed Jalali, Gjergji Kasneci

Responsive image

Auto-TLDR; A novel approach for aggregating the Gaussian experts by detecting strong violations of conditional independence

Slides Poster Similar

Distributed Gaussian processes (DGPs) are prominent local approximation methods to scale Gaussian processes (GPs) to large datasets. Instead of a global estimation, they train local experts by dividing the training set into subsets, thus reducing the time complexity. This strategy is based on the conditional independence assumption, which basically means that there is a perfect diversity between the local experts. In practice, however, this assumption is often violated, and the aggregation of experts leads to sub-optimal and inconsistent solutions. In this paper, we propose a novel approach for aggregating the Gaussian experts by detecting strong violations of conditional independence. The dependency between experts is determined by using a Gaussian graphical model, which yields the precision matrix. The precision matrix encodes conditional dependencies between experts and is used to detect strongly dependent experts and construct an improved aggregation. Using both synthetic and real datasets, our experimental evaluations illustrate that our new method outperforms other state-of-the-art (SOTA) DGP approaches while being substantially more time-efficient than SOTA approaches, which build on independent experts.

Multi-Modal Deep Clustering: Unsupervised Partitioning of Images

Guy Shiran, Daphna Weinshall

Responsive image

Auto-TLDR; Multi-Modal Deep Clustering for Unlabeled Images

Slides Poster Similar

The clustering of unlabeled raw images is a daunting task, which has recently been approached with some success by deep learning methods. Here we propose an unsupervised clustering framework, which learns a deep neural network in an end-to-end fashion, providing direct cluster assignments of images without additional processing. Multi-Modal Deep Clustering (MMDC), trains a deep network to align its image embeddings with target points sampled from a Gaussian Mixture Model distribution. The cluster assignments are then determined by mixture component association of image embeddings. Simultaneously, the same deep network is trained to solve an additional self-supervised task. This pushes the network to learn more meaningful image representations and stabilizes the training. Experimental results show that MMDC achieves or exceeds state-of-the-art performance on four challenging benchmarks. On natural image datasets we improve on previous results with significant margins of up to 11% absolute accuracy points, yielding an accuracy of 70% on CIFAR-10 and 61% on STL-10.

Adaptive Matching of Kernel Means

Miao Cheng, Xinge You

Responsive image

Auto-TLDR; Adaptive Matching of Kernel Means for Knowledge Discovery and Feature Learning

Slides Poster Similar

As a promising step, the performance of data analysis and feature learning are able to be improved if certain pattern matching mechanism is available. One of the feasible solutions can refer to the importance estimation of instances, and consequently, kernel mean matching (KMM) has become an important method for knowledge discovery and novelty detection in general. Furthermore, the existing KMM methods have focused on concrete learning frameworks. In this work, a novel approach to adaptive matching of kernel means is proposed, and selected data with high importance are adopted to achieve calculation efficiency with optimization. In addition, scalable learning can be conducted in proposed method as a generalized solution with appended data. The experimental results on a wide variety of real-world data sets demonstrate the proposed method is able to give outstanding performance compared with several state-of-the-art methods, while calculation efficiency can be preserved.

Feature Extraction and Selection Via Robust Discriminant Analysis and Class Sparsity

Ahmad Khoder, Fadi Dornaika

Responsive image

Auto-TLDR; Hybrid Linear Discriminant Embedding for supervised multi-class classification

Slides Poster Similar

The main goal of discriminant embedding is to extract features that can be compact and informative representations of the original set of features. This paper introduces a hybrid scheme for linear feature extraction for supervised multi-class classification. We introduce a unifying criterion that is able to retain the advantages of robust sparse LDA and Inter-class sparsity. Thus, the estimated transformation includes two types of discrimination which are the inter-class sparsity and robust Linear Discriminant Analysis with feature selection. In order to optimize the proposed objective function, we deploy an iterative alternating minimization scheme for estimating the linear transformation and the orthogonal matrix. The introduced scheme is generic in the sense that it can be used for combining and tuning many other linear embedding methods. In the lights of the experiments conducted on six image datasets including faces, objects, and digits, the proposed scheme was able to outperform competing methods in most of the cases.

On Learning Random Forests for Random Forest Clustering

Manuele Bicego, Francisco Escolano

Responsive image

Auto-TLDR; Learning Random Forests for Clustering

Slides Poster Similar

In this paper we study the poorly investigated problem of learning Random Forests for distance-based Random Forest clustering. We studied both classic schemes as well as alternative approaches, novel in this context. In particular, we investigated the suitability of Gaussian Density Forests, Random Forests specifically designed for density estimation. Further, we introduce a novel variant of Random Forest, based on an effective non parametric by-pass estimator of the Renyi entropy, which can be useful when the parametric assumption is too strict. An empirical evaluation involving different datasets and different RF-clustering strategies confirms that the learning step is crucial for RF-clustering. We also present a set of practical guidelines useful to determine the most suitable variant of RF-clustering according to the problem under examination.

Nonlinear Ranking Loss on Riemannian Potato Embedding

Byung Hyung Kim, Yoonje Suh, Honggu Lee, Sungho Jo

Responsive image

Auto-TLDR; Riemannian Potato for Rank-based Metric Learning

Slides Poster Similar

We propose a rank-based metric learning method by leveraging a concept of the Riemannian Potato for better separating non-linear data. By exploring the geometric properties of Riemannian manifolds, the proposed loss function optimizes the measure of dispersion using the distribution of Riemannian distances between a reference sample and neighbors and builds a ranked list according to the similarities. We show the proposed function can learn a hypersphere for each class, preserving the similarity structure inside it on Riemannian manifold. As a result, compared with Euclidean distance-based metric, our method can further jointly reduce the intra-class distances and enlarge the inter-class distances for learned features, consistently outperforming state-of-the-art methods on three widely used non-linear datasets.

Improved Deep Classwise Hashing with Centers Similarity Learning for Image Retrieval

Ming Zhang, Hong Yan

Responsive image

Auto-TLDR; Deep Classwise Hashing for Image Retrieval Using Center Similarity Learning

Slides Poster Similar

Deep supervised hashing for image retrieval has attracted researchers' attention due to its high efficiency and superior retrieval performance. Most existing deep supervised hashing works, which are based on pairwise/triplet labels, suffer from the expensive computational cost and insufficient utilization of the semantics information. Recently, deep classwise hashing introduced a classwise loss supervised by class labels information alternatively; however, we find it still has its drawback. In this paper, we propose an improved deep classwise hashing, which enables hashing learning and class centers learning simultaneously. Specifically, we design a two-step strategy on center similarity learning. It interacts with the classwise loss to attract the class center to concentrate on the intra-class samples while pushing other class centers as far as possible. The centers similarity learning contributes to generating more compact and discriminative hashing codes. We conduct experiments on three benchmark datasets. It shows that the proposed method effectively surpasses the original method and outperforms state-of-the-art baselines under various commonly-used evaluation metrics for image retrieval.

Unveiling Groups of Related Tasks in Multi-Task Learning

Jordan Frecon, Saverio Salzo, Massimiliano Pontil

Responsive image

Auto-TLDR; Continuous Bilevel Optimization for Multi-Task Learning

Slides Poster Similar

A common approach in multi-task learning is to encourage the tasks to share a low dimensional representation. This has led to the popular method of trace norm regularization, which has proved effective in many applications. In this paper, we extend this approach by allowing the tasks to partition into different groups, within which trace norm regularization is separately applied. We propose a continuous bilevel optimization framework to simultaneously identify groups of related tasks and learn a low dimensional representation within each group. Hinging on recent results on the derivative of generalized matrix functions, we devise a smooth approximation of the upper-level objective via a dual forward-backward algorithm with Bregman distances. This allows us to solve the bilevel problem by a gradient-based scheme. Numerical experiments on synthetic and benchmark datasets support the effectiveness of the proposed method.

Learning Sign-Constrained Support Vector Machines

Kenya Tajima, Kouhei Tsuchida, Esmeraldo Ronnie Rey Zara, Naoya Ohta, Tsuyoshi Kato

Responsive image

Auto-TLDR; Constrained Sign Constraints for Learning Linear Support Vector Machine

Poster Similar

Domain knowledge is useful to improve the generalization performance of learning machines. Sign constraints are a handy representation to combine domain knowledge with learning machine. In this paper, we consider constraining the signs of the weight coefficients in learning the linear support vector machine, and develop two optimization algorithms for minimizing the empirical risk under the sign constraints. One of the two algorithms is based on the projected gradient method, in which each iteration of the projected gradient method takes O(nd) computational cost and the sublinear convergence of the objective error is guaranteed. The second algorithm is based on the Frank-Wolfe method that also converges sublinearly and possesses a clear termination criterion. We show that each iteration of the Frank-Wolfe also requires O(nd) cost. Furthermore, we derive the explicit expression for the minimal iteration number to ensure an epsilon-accurate solution by analyzing the curvature of the objective function. Finally, we empirically demonstrate that the sign constraints are a promising technique when similarities to the training examples compose the feature vector.

SSDL: Self-Supervised Domain Learning for Improved Face Recognition

Samadhi Poornima Kumarasinghe Wickrama Arachchilage, Ebroul Izquierdo

Responsive image

Auto-TLDR; Self-supervised Domain Learning for Face Recognition in unconstrained environments

Slides Poster Similar

Face recognition in unconstrained environments is challenging due to variations in illumination, quality of sensing, motion blur and etc. An individual’s face appearance can vary drastically under different conditions creating a gap between train (source) and varying test (target) data. The domain gap could cause decreased performance levels in direct knowledge transfer from source to target. Despite fine-tuning with domain specific data could be an effective solution, collecting and annotating data for all domains is extremely expensive. To this end, we propose a self-supervised domain learning (SSDL) scheme that trains on triplets mined from unlabelled data. A key factor in effective discriminative learning, is selecting informative triplets. Building on most confident predictions, we follow an “easy-to-hard” scheme of alternate triplet mining and self-learning. Comprehensive experiments on four different benchmarks show that SSDL generalizes well on different domains.

Learning Stable Deep Predictive Coding Networks with Weight Norm Supervision

Guo Ruohao

Responsive image

Auto-TLDR; Stability of Predictive Coding Network with Weight Norm Supervision

Slides Poster Similar

Predictive Coding Network (PCN) is an important neural network inspired by visual processing models in neuroscience. It combines the feedforward and feedback processing and has the architecture of recurrent neural networks (RNNs). This type of network is usually trained with backpropagation through time (BPTT). With infinite recurrent steps, PCN is a dynamic system. However, as one of the most important properties, stability is rarely studied in this type of network. Inspired by reservoir computing, we investigate the stability of hierarchical RNNs from the perspective of dynamic systems, and propose a sufficient condition for their echo state property (ESP). Our study shows the global stability is determined by stability of the local layers and the feedback between neighboring layers. Based on it, we further propose Weight Norm Supervision, a new algorithm that controls the stability of PCN dynamics by imposing different weight norm constraints on different parts of the network. We compare our approach with other training methods in terms of stability and prediction capability. The experiments show that our algorithm learns stable PCNs with a reliable prediction precision in the most effective and controllable way.

Learning Sparse Deep Neural Networks Using Efficient Structured Projections on Convex Constraints for Green AI

Michel Barlaud, Frederic Guyard

Responsive image

Auto-TLDR; Constrained Deep Neural Network with Constrained Splitting Projection

Slides Poster Similar

In recent years, deep neural networks (DNN) have been applied to different domains and achieved dramatic performance improvements over state-of-the-art classical methods. These performances of DNNs were however often obtained with networks containing millions of parameters and which training required heavy computational power. In order to cope with this computational issue a huge literature deals with proximal regularization methods which are time consuming.\\ In this paper, we propose instead a constrained approach. We provide the general framework for our new splitting projection gradient method. Our splitting algorithm iterates a gradient step and a projection on convex sets. We study algorithms for different constraints: the classical $\ell_1$ unstructured constraint and structured constraints such as the nuclear norm, the $\ell_{2,1} $ constraint (Group LASSO). We propose a new $\ell_{1,1} $ structured constraint for which we provide a new projection algorithm We demonstrate the effectiveness of our method on three popular datasets (MNIST, Fashion MNIST and CIFAR). Experiments on these datasets show that our splitting projection method with our new $\ell_{1,1} $ structured constraint provides the best reduction of memory and computational power. Experiments show that fully connected linear DNN are more efficient for green AI.

A Unified Framework for Distance-Aware Domain Adaptation

Fei Wang, Youdong Ding, Huan Liang, Yuzhen Gao, Wenqi Che

Responsive image

Auto-TLDR; distance-aware domain adaptation

Slides Poster Similar

Unsupervised domain adaptation has achieved significant results by leveraging knowledge from a source domain to learn a related but unlabeled target domain. Previous methods are insufficient to model domain discrepancy and class discrepancy, which may lead to misalignment and poor adaptation performance. To address this problem, in this paper, we propose a unified framework, called distance-aware domain adaptation, which is fully aware of both cross-domain distance and class-discriminative distance. In addition, second-order statistics distance and manifold alignment are also exploited to extract more information from data. In this manner, the generalization error of the target domain in classification problems can be reduced substantially. To validate the proposed method, we conducted experiments on five public datasets and an ablation study. The results demonstrate the good performance of our proposed method.

Supervised Feature Embedding for Classification by Learning Rank-Based Neighborhoods

Ghazaal Sheikhi, Hakan Altincay

Responsive image

Auto-TLDR; Supervised Feature Embedding with Representation Learning of Rank-based Neighborhoods

Slides Similar

In feature embedding, the recovery of associated discriminative information in the reduced subspace is critical for downstream classifiers. In this study, a supervised feature embedding method is proposed inspired by the well-known word embedding technique, word2vec. Proposed embedding method is implemented as representative learning of rank-based neighborhoods. The notion of context words in word2vec is extended into neighboring instances within a given window. Neighborship is defined using ranks of instances rather than their values so that regions with different densities are captured properly. Each sample is represented by a unique one-hot vector whereas its neighbors are encoded by several two-hot vectors. The two-hot vectors are identical for neighboring samples of the same class. A feed-forward neural network with a continuous projection layer, then learns the mapping from one-hot vectors to multiple two-hot vectors. The hidden layer determines the reduced subspace for the train samples. The obtained transformation is then applied on test data to find a lower-dimensional representation. Proposed method is tested in classification problems on 10 UCI data sets. Experimental results confirm that the proposed method is effective in finding a discriminative representation of the features and outperforms several supervised embedding approaches in terms of classification performance.

Adversarial Encoder-Multi-Task-Decoder for Multi-Stage Processes

Andre Mendes, Julian Togelius, Leandro Dos Santos Coelho

Responsive image

Auto-TLDR; Multi-Task Learning and Semi-Supervised Learning for Multi-Stage Processes

Similar

In multi-stage processes, decisions occur in an ordered sequence of stages. Early stages usually have more observations with general information (easier/cheaper to collect), while later stages have fewer observations but more specific data. This situation can be represented by a dual funnel structure, in which the sample size decreases from one stage to the other while the information increases. Training classifiers in this scenario is challenging since information in the early stages may not contain distinct patterns to learn (underfitting). In contrast, the small sample size in later stages can cause overfitting. We address both cases by introducing a framework that combines adversarial autoencoders (AAE), multi-task learning (MTL), and multi-label semi-supervised learning (MLSSL). We improve the decoder of the AAE with an MTL component so it can jointly reconstruct the original input and use feature nets to predict the features for the next stages. We also introduce a sequence constraint in the output of an MLSSL classifier to guarantee the sequential pattern in the predictions. Using real-world data from different domains (selection process, medical diagnosis), we show that our approach outperforms other state-of-the-art methods.

Uniform and Non-Uniform Sampling Methods for Sub-Linear Time K-Means Clustering

Yuanhang Ren, Ye Du

Responsive image

Auto-TLDR; Sub-linear Time Clustering with Constant Approximation Ratio for K-Means Problem

Slides Poster Similar

The $k$-means problem is arguably the most well-known clustering problem in machine learning, and lots of approximation algorithms have been proposed for it. However, many of these algorithms may become infeasible when data is huge. Sub-linear time algorithms with constant approximation ratios are desirable in this scenario. In this paper, we first improve the analysis of the algorithm proposed by \cite{Mohan:2017:BNA:3172077.3172235} by sharpening the approximation ratio from $4(\alpha+\beta)$ to $\alpha+\beta$. Moreover, on mild assumptions of the data, a constant approximation ratio can be achieved in poly-logarithmic time by the algorithm. Furthermore, we propose a novel sub-linear time clustering algorithm called {\it Double-K-M$\text{C}^2$ sampling} as well. Experiments on the data clustering task and the image segmentation task have validated the effectiveness of our algorithms.

A Multilinear Sampling Algorithm to Estimate Shapley Values

Ramin Okhrati, Aldo Lipani

Responsive image

Auto-TLDR; A sampling method for Shapley values for multilayer Perceptrons

Slides Poster Similar

Shapley values are great analytical tools in game theory to measure the importance of a player in a game. Due to their axiomatic and desirable properties such as efficiency, they have become popular for feature importance analysis in data science and machine learning. However, the time complexity to compute Shapley values based on the original formula is exponential, and as the number of features increases, this becomes infeasible. Castro et al. [1] developed a sampling algorithm, to estimate Shapley values. In this work, we propose a new sampling method based on a multilinear extension technique as applied in game theory. The aim is to provide a more efficient (sampling) method for estimating Shapley values. Our method is applicable to any machine learning model, in particular for either multiclass classifications or regression problems. We apply the method to estimate Shapley values for multilayer Perceptrons (MLPs) and through experimentation on two datasets, we demonstrate that our method provides more accurate estimations of the Shapley values by reducing the variance of the sampling statistics

Active Sampling for Pairwise Comparisons via Approximate Message Passing and Information Gain Maximization

Aliaksei Mikhailiuk, Clifford Wilmot, Maria Perez-Ortiz, Dingcheng Yue, Rafal Mantiuk

Responsive image

Auto-TLDR; ASAP: An Active Sampling Algorithm for Pairwise Comparison Data

Slides Similar

Pairwise comparison data arise in many domains with subjective assessment experiments, for example in image and video quality assessment. In these experiments observers are asked to express a preference between two conditions. However, many pairwise comparison protocols require a large number of comparisons to infer accurate scores, which may be unfeasible when each comparison is time-consuming (e.g. videos) or expensive (e.g. medical imaging). This motivates the use of an active sampling algorithm that chooses only the most informative pairs for comparison. In this paper we propose ASAP, an active sampling algorithm based on approximate message passing and expected information gain maximization. Unlike most existing methods, which rely on partial updates of the posterior distribution, we are able to perform full updates and therefore much improve the accuracy of the inferred scores. The algorithm relies on three techniques for reducing computational cost: inference based on approximate message passing, selective evaluations of the information gain, and selecting pairs in a batch that forms a minimum spanning tree of the inverse of information gain. We demonstrate, with real and synthetic data, that ASAP offers the highest accuracy of inferred scores compared to the existing methods. We also provide an open-source GPU implementation of ASAP for large-scale experiments.

Nearest Neighbor Classification Based on Activation Space of Convolutional Neural Network

Xinbo Ju, Shuo Shao, Huan Long, Weizhe Wang

Responsive image

Auto-TLDR; Convolutional Neural Network with Convex Hull Based Classifier

Poster Similar

In this paper, we propose a new image classifier based on the incorporation of the nearest neighbor algorithm and the activation space of convolutional neural network. The classifier has been successfully used on some state-of-the-art models and further improve their performance. Main technique tools we used are convex hull based classification and its acceleration. We find that 1) in several cases, the classifier can reach higher accuracy than original CNN; 2) by sampling, the classifier can work more efficiently; 3) centroid of each convex hull shows surprising ability in classification. Most of the work has strong geometry meanings, which helps us have a new understanding about convolutional layers.

Supervised Domain Adaptation Using Graph Embedding

Lukas Hedegaard, Omar Ali Sheikh-Omar, Alexandros Iosifidis

Responsive image

Auto-TLDR; Domain Adaptation from the Perspective of Multi-view Graph Embedding and Dimensionality Reduction

Slides Poster Similar

Getting deep convolutional neural networks to perform well requires a large amount of training data. When the available labelled data is small, it is often beneficial to use transfer learning to leverage a related larger dataset (source) in order to improve the performance on the small dataset (target). Among the transfer learning approaches, domain adaptation methods assume that distributions between the two domains are shifted and attempt to realign them. In this paper, we consider the domain adaptation problem from the perspective of multi-view graph embedding and dimensionality reduction. Instead of solving the generalised eigenvalue problem to perform the embedding, we formulate the graph-preserving criterion as loss in the neural network and learn a domain-invariant feature transformation in an end-to-end fashion. We show that the proposed approach leads to a powerful Domain Adaptation framework which generalises the prior methods CCSA and d-SNE, and enables simple and effective loss designs; an LDA-inspired instantiation of the framework leads to performance on par with the state-of-the-art on the most widely used Domain Adaptation benchmarks, Office31 and MNIST to USPS datasets.