ICPR2020 Paper Browser

Paper download is intended for registered attendees only, and is subjected to the IEEE Copyright Policy. Any other use is strongly forbidden.

Unveiling Groups of Related Tasks in Multi-Task Learning

Jordan Frecon, Saverio Salzo, Massimiliano Pontil

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

Abstract Slides Poster

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.

Similar papers

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

Michel Barlaud, Antonin Chambolle, Jean_Baptiste Caillau

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

Abstract 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.

Temporal Pattern Detection in Time-Varying Graphical Models

Federico Tomasi, Veronica Tozzo, Annalisa Barla

Auto-TLDR; A dynamical network inference model that leverages on kernels to consider general temporal patterns

Abstract Slides Poster Similar

Graphical models allow to describe the interplay among variables of a system through a compact representation, suitable when relations evolve over time. For example, in a biological setting, genes interact differently depending on external environmental or metabolic factors. To incorporate this dynamics a viable strategy is to estimate a sequence of temporally related graphs assuming similarity among samples in different time points. While adjacent time points may direct the analysis towards a robust estimate of the underlying graph, the resulting model will not incorporate long-term or recurrent temporal relationships. In this work we propose a dynamical network inference model that leverages on kernels to consider general temporal patterns (such as circadian rhythms or seasonality). We show how our approach may also be exploited when the recurrent patterns are unknown, by coupling the network inference with a clustering procedure that detects possibly non-consecutive similar networks. Such clusters are then used to build similarity kernels. The convexity of the functional is determined by whether we impose or infer the kernel. In the first case, the optimisation algorithm exploits efficiently proximity operators with closed-form solutions. In the other case, we resort to an alternating minimisation procedure which jointly learns the temporal kernel and the underlying network. Extensive analysis on synthetic data shows the efficacy of our models compared to state-of-the-art methods. Finally, we applied our approach on two real-world applications to show how considering long-term patterns is fundamental to have insights on the behaviour of a complex system.

Learning Sign-Constrained Support Vector Machines

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

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

Abstract 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.

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

Michel Barlaud, Frederic Guyard

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

Abstract 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.

Low-Cost Lipschitz-Independent Adaptive Importance Sampling of Stochastic Gradients

Huikang Liu, Xiaolu Wang, Jiajin Li, Man-Cho Anthony So

Auto-TLDR; Adaptive Importance Sampling for Stochastic Gradient Descent

Abstract Slides Similar

Stochastic gradient descent (SGD) usually samples training data based on the uniform distribution, which may not be a good choice because of the high variance of its stochastic gradient. Thus, importance sampling methods are considered in the literature to improve the performance. Most previous work on SGD-based methods with importance sampling requires the knowledge of Lipschitz constants of all component gradients, which are in general difficult to estimate. In this paper, we study an adaptive importance sampling method for common SGD-based methods by exploiting the local first-order information without knowing any Lipschitz constants. In particular, we periodically changes the sampling distribution by only utilizing the gradient norms in the past few iterations. We prove that our adaptive importance sampling non-asymptotically reduces the variance of the stochastic gradients in SGD, and thus better convergence bounds than that for vanilla SGD can be obtained. We extend this sampling method to several other widely used stochastic gradient algorithms including SGD with momentum and ADAM. Experiments on common convex learning problems and deep neural networks illustrate notably enhanced performance using the adaptive sampling strategy.

RNN Training along Locally Optimal Trajectories via Frank-Wolfe Algorithm

Yun Yue, Ming Li, Venkatesh Saligrama, Ziming Zhang

Auto-TLDR; Frank-Wolfe Algorithm for Efficient Training of RNNs

Abstract Slides Poster Similar

We propose a novel and efficient training method for RNNs by iteratively seeking a local minima on the loss surface within a small region, and leverage this directional vector for the update, in an outer-loop. We propose to utilize the Frank-Wolfe (FW) algorithm in this context. Although, FW implicitly involves normalized gradients, which can lead to a slow convergence rate, we develop a novel RNN training method that, surprisingly, even with the additional cost, the overall training cost is empirically observed to be lower than back-propagation. Our method leads to a new Frank-Wolfe method, that is in essence an SGD algorithm with a restart scheme. We prove that under certain conditions our algorithm has a sublinear convergence rate of $O(1/\epsilon)$ for $\epsilon$ error. We then conduct empirical experiments on several benchmark datasets including those that exhibit long-term dependencies, and show significant performance improvement. We also experiment with deep RNN architectures and show efficient training performance. Finally, we demonstrate that our training method is robust to noisy data.

Feature Extraction and Selection Via Robust Discriminant Analysis and Class Sparsity

Ahmad Khoder, Fadi Dornaika

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

Abstract 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.

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

Fadi Dornaika, Ahmad Khoder

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

Abstract 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.

Stochastic Runge-Kutta Methods and Adaptive SGD-G2 Stochastic Gradient Descent

Gabriel Turinici, Imen Ayadi

Auto-TLDR; Adaptive Stochastic Runge Kutta for the Minimization of the Loss Function

Abstract Slides Poster Similar

The minimization of the loss function is of paramount importance in deep neural networks. Many popular optimization algorithms have been shown to correspond to some evolution equation of gradient flow type. Inspired by the numerical schemes used for general evolution equations, we introduce a second-order stochastic Runge Kutta method and show that it yields a consistent procedure for the minimization of the loss function. In addition, it can be coupled, in an adaptive framework, with the Stochastic Gradient Descent (SGD) to adjust automatically the learning rate of the SGD The resulting adaptive SGD, called SGD-G2, shows good results in terms of convergence speed when tested on standard data-sets.

T-SVD Based Non-Convex Tensor Completion and Robust Principal Component Analysis

Tao Li, Jinwen Ma

Auto-TLDR; Non-Convex tensor rank surrogate function and non-convex sparsity measure for tensor recovery

Abstract Slides Poster Similar

In this paper, we propose a novel non-convex tensor rank surrogate function and a novel non-convex sparsity measure. The basic idea is to sidestep the bias of $\ell_1-$norm by introducing the concavity. Furthermore, we employ this non-convex penalty in tensor recovery problems such as tensor completion and tensor robust principal component analysis. Due to the concavity, the parameters of these models are difficult to solve. To tackle this problem, we devise a majorization minimization algorithm that can optimize the upper bound of the original function in each iteration, and every sub-problem is solved by the alternating direction multiplier method. We also analyze the theoretical properties of the proposed algorithm. Finally, the experimental results on natural and hyperspectral images demonstrate the efficacy and efficiency of the proposed method.

Soft Label and Discriminant Embedding Estimation for Semi-Supervised Classification

Fadi Dornaika, Abdullah Baradaaji, Youssof El Traboulsi

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

Abstract 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.

Fast Subspace Clustering Based on the Kronecker Product

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

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

Abstract 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.

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

Zhan Wang, Lizhi Wang, Hua Huang

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

Abstract 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.

A Multilinear Sampling Algorithm to Estimate Shapley Values

Ramin Okhrati, Aldo Lipani

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

Abstract 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

Subspace Clustering Via Joint Unsupervised Feature Selection

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

Auto-TLDR; Unsupervised Feature Selection for Subspace Clustering

Abstract 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.

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

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

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

Abstract 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.

Scalable Direction-Search-Based Approach to Subspace Clustering

Yicong He, George Atia

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

Abstract 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.

An Efficient Empirical Solver for Localized Multiple Kernel Learning Via DNNs

Ziming Zhang

Auto-TLDR; Localized Multiple Kernel Learning using LMKL-Net

Abstract Slides Poster Similar

In this paper we propose solving localized multiple kernel learning (LMKL) using LMKL-Net, a feedforward deep neural network (DNN). In contrast to previous works, as a learning principle we propose parameterizing the gating function for learning kernel combination weights and the multiclass classifier using an attentional network (AN) and a multilayer perceptron (MLP), respectively. Such interpretability helps us better understand how the network solves the problem. Thanks to stochastic gradient descent (SGD), our approach has {\em linear} computational complexity in training. Empirically on benchmark datasets we demonstrate that with comparable or better accuracy than the state-of-the-art, our LMKL-Net can be trained about {\bf two orders of magnitude} faster with about {\bf two orders of magnitude} smaller memory footprint for large-scale learning.

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

Andre Mendes, Julian Togelius, Leandro Dos Santos Coelho

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

Abstract 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.

Aggregating Dependent Gaussian Experts in Local Approximation

Hamed Jalali, Gjergji Kasneci

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

Abstract 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.

Sketch-Based Community Detection Via Representative Node Sampling

Mahlagha Sedghi, Andre Beckus, George Atia

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

Abstract 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.

3CS Algorithm for Efficient Gaussian Process Model Retrieval

Fabian Berns, Kjeld Schmidt, Ingolf Bracht, Christian Beecks

Auto-TLDR; Efficient retrieval of Gaussian Process Models for large-scale data using divide-&-conquer-based approach

Abstract Slides Poster Similar

Gaussian Process Models (GPMs) have been applied for various pattern recognition tasks due to their analytical tractability, ability to quantify uncertainty for their own results as well as to subsume prominent other regression techniques. Despite these promising prospects their super-quadratic computation time complexity for model selection and evaluation impedes its broader application for more than a few thousand data points. Although there have been many proposals towards Gaussian Processes for large-scale data, those only offer a linearly scaling improvement to a cubical scaling problem. In particular, solutions like the Nystrom approximation or sparse matrices are only taking fractions of the given data into account and subsequently lead to inaccurate models. In this paper, we thus propose a divide-&-conquer-based approach, that allows to efficiently retrieve GPMs for large-scale data. The resulting model is composed of independent pattern representations for non-overlapping segments of the given data and consequently reduces computation time significantly. Our performance analysis indicates that our proposal is able to outperform state-of-the-art algorithms for GPM retrieval with respect to the qualities of efficiency and accuracy.

Supervised Domain Adaptation Using Graph Embedding

Lukas Hedegaard, Omar Ali Sheikh-Omar, Alexandros Iosifidis

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

Abstract 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.

Automatically Mining Relevant Variable Interactions Via Sparse Bayesian Learning

Ryoichiro Yafune, Daisuke Sakuma, Yasuo Tabei, Noritaka Saito, Hiroto Saigo

Auto-TLDR; Sparse Bayes for Interpretable Non-linear Prediction

Abstract Slides Poster Similar

With the rapid increase in the availability of large amount of data, prediction is becoming increasingly popular, and has widespread through our daily life. However, powerful non- linear prediction methods such as deep learning and SVM suffer from interpretability problem, making it hard to use in domains where the reason for decision making is required. In this paper, we develop an interpretable non-linear model called itemset Sparse Bayes (iSB), which builds a Bayesian probabilistic model, while simultaneously considering variable interactions. In order to suppress the resulting large number of variables, sparsity is imposed on regression weights by a sparsity inducing prior. As a subroutine to search for variable interactions, itemset enumeration algorithm is employed with a novel bounding condition. In computational experiments using real-world dataset, the proposed method performed better than decision tree by 10% in terms of r-squared . We also demonstrated the advantage of our method in Bayesian optimization setting, in which the proposed approach could successfully find the maximum of an unknown function faster than Gaussian process. The interpretability of iSB is naturally inherited to Bayesian optimization, thereby gives us a clue to understand which variables interactions are important in optimizing an unknown function.

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

Sayantan Mitra, Sriparna Saha

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

Abstract 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.

A Novel Random Forest Dissimilarity Measure for Multi-View Learning

Hongliu Cao, Simon Bernard, Robert Sabourin, Laurent Heutte

Auto-TLDR; Multi-view Learning with Random Forest Relation Measure and Instance Hardness

Abstract Slides Poster Similar

Multi-view learning is a learning task in which data is described by several concurrent representations. Its main challenge is most often to exploit the complementarities between these representations to help solve a classification/regression task. This is a challenge that can be met nowadays if there is a large amount of data available for learning. However, this is not necessarily true for all real-world problems, where data are sometimes scarce (e.g. problems related to the medical environment). In these situations, an effective strategy is to use intermediate representations based on the dissimilarities between instances. This work presents new ways of constructing these dissimilarity representations, learning them from data with Random Forest classifiers. More precisely, two methods are proposed, which modify the Random Forest proximity measure, to adapt it to the context of High Dimension Low Sample Size (HDLSS) multi-view classification problems. The second method, based on an Instance Hardness measurement, is significantly more accurate than other state-of-the-art measurements including the original RF Proximity measurement and the Large Margin Nearest Neighbor (LMNN) metric learning measurement.

A New Convex Loss Function for Multiple Instance Support Vector Machines

Sang-Baeg Kim, Jung-Man Bae

Auto-TLDR; WR-SVM: A Novel Multiple Instance SVM for Video Classification

Abstract Slides Poster Similar

We develop a novel multiple instance SVM based on maximizing the minimum witness rate (WR) among positive bags. We solve this nonlinear integer programming problem by the multiple instance SVM formulation with a convex loss function, where unknown integer labels of instances in positive bags are relaxed to real numbers between -1 and 1 using the tanh(.) to estimate WR of a positive bag. Our new model, WR-SVM, also can incorporates the multiple instance learning (MIL) problem into a simple deep neural network framework with no additional MIL pooling layers. WR-SVM is robust to input perturbation by eliminating the imbalance between positive instances allocated to positive bags. A better generalization ability is expected by the large margin due to the balanced allocation of positive instances to positive bags. We perform experiments on various video datasets to verify the effectiveness of our method for video classification. The results of WR-SVM outperform the state-of-the-art for MIL-based video classification models.

Bayesian Active Learning for Maximal Information Gain on Model Parameters

Kasra Arnavaz, Aasa Feragen, Oswin Krause, Marco Loog

Auto-TLDR; Bayesian assumptions for Bayesian classification

Abstract Slides Poster Similar

The fact that machine learning models, despite their advancements, are still trained on randomly gathered data is proof that a lasting solution to the problem of optimal data gathering has not yet been found. In this paper, we investigate whether a Bayesian approach to the classification problem can provide assumptions under which one is guaranteed to perform at least as good as random sampling. For a logistic regression model, we show that maximal expected information gain on model parameters is a promising criterion for selecting samples, assuming that our classification model is well-matched to the data. Our derived criterion is closely related to the maximum model change. We experiment with data sets which satisfy this assumption to varying degrees to see how sensitive our performance is to the violation of our assumption in practice.

Mean Decision Rules Method with Smart Sampling for Fast Large-Scale Binary SVM Classification

Alexandra Makarova, Mikhail Kurbakov, Valentina Sulimova

Auto-TLDR; Improving Mean Decision Rule for Large-Scale Binary SVM Problems

Abstract Slides Poster Similar

This paper relies on the Mean Decision Rule (MDR) method for solving large-scale binary SVM problems. It consists in taking small random samples of the full dataset and separate training for each of them with consecutive averaging the respective individual decision rules to obtain a final one. This paper proposes two new approaches to improve it. The first proposed approach is a new sampling technique that exploits SVM and MDR properties to fast form so called smart samples by selecting only the objects, that are candidates to be the support ones. The proposed technique essentially increases MDR convergence and allows to reach the highest quality in less time. In the case of kernel-based MDR (KMDR) the proposed sampling technique allows additionally to reduce the number of support objects in the final decision rule and, as a result, to decrease the recognition time. The second proposed approach is a new data strategy to accelerate random access to large datasets stored in the traditional libsvm format. The proposed strategy allows to quickly extract random subsets of objects from a file and load them into RAM, and is it also suitable for any sampling-based methods, including stochastic gradient methods. Joint using of the proposed approaches with (K)MDR allows to obtain the best (or near the best) decision of large-scale binary SVM problems faster, compared to the existing SVM solvers.

An Intransitivity Model for Matchup and Pairwise Comparison

Yan Gu, Jiuding Duan, Hisashi Kashima

Auto-TLDR; Blade-Chest: A Low-Rank Matrix Approach for Probabilistic Ranking of Players

Abstract Slides Poster Similar

Ranking is a ubiquitous problem appearing in many real-world applications. The superior players or objects are oftentimes determined by a matchup or pairwise comparison. Various models have been developed to integrate the matchup results into a single ranking list of players and to further predict the results of future matchups. Amongst them, the Bradley-Terry model is a mainstream model that achieves the goals by constructing explicit probabilistic interpretation. However, the model suffers from its strong assumption of transitive relationships and becomes vulnerable in practices where intransitive relationships exist. Blade-Chest model is an alternative solution to this intransitivity challenge by allowing the multi-dimensional representation of players. In this paper, we propose a low-rank matrix approach to characterize all players and generalize the related works by introducing a unified framework. Our experimental results on synthetic datasets and real-world datasets show that the proposed model is stably competitive with the standard models in terms of the consistency of probabilistic model interpretation and the predictive performance in out-of-sample tests.

Learning Stable Deep Predictive Coding Networks with Weight Norm Supervision

Guo Ruohao

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

Abstract 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.

On the Global Self-attention Mechanism for Graph Convolutional Networks

Chen Wang, Deng Chengyuan

Auto-TLDR; Global Self-Attention Mechanism for Graph Convolutional Networks

Abstract Slides Similar

Applying Global Self-Attention (GSA) mechanism over features has achieved remarkable success on Convolutional Neural Networks (CNNs). However, it is not clear if Graph Convolutional Networks (GCNs) can similarly benefit from such a technique. In this paper, inspired by the similarity between CNNs and GCNs, we study the impact of the Global Self-Attention mechanism on GCNs. We find that consistent with the intuition, the GSA mechanism allows GCNs to capture feature-based vertex relations regardless of edge connections; As a result, the GSA mechanism can introduce extra expressive power to the GCNs. Furthermore, we analyze the impacts of the GSA mechanism on the issues of overfitting and over-smoothing. We prove that the GSA mechanism can alleviate both the overfitting and the over-smoothing issues based on some recent technical developments. Experiments on multiple benchmark datasets illustrate both superior expressive power and less significant overfitting and over-smoothing problems for the GSA-augmented GCNs, which corroborate the intuitions and the theoretical results.

A Spectral Clustering on Grassmann Manifold Via Double Low Rank Constraint

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

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

Abstract 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.

Double Manifolds Regularized Non-Negative Matrix Factorization for Data Representation

Jipeng Guo, Shuai Yin, Yanfeng Sun, Yongli Hu

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

Abstract 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.

Deep Convolutional Embedding for Digitized Painting Clustering

Giovanna Castellano, Gennaro Vessio

Auto-TLDR; A Deep Convolutional Embedding Model for Clustering Artworks

Abstract Slides Poster Similar

Clustering artworks is difficult because of several reasons. On one hand, recognizing meaningful patterns in accordance with domain knowledge and visual perception is extremely hard. On the other hand, the application of traditional clustering and feature reduction techniques to the highly dimensional pixel space can be ineffective. To address these issues, we propose to use a deep convolutional embedding model for digitized painting clustering, in which the task of mapping the input raw data to an abstract, latent space is jointly optimized with the task of finding a set of cluster centroids in this latent feature space. Quantitative and qualitative experimental results show the effectiveness of the proposed method. The model is also able to outperform other state-of-the-art deep clustering approaches to the same problem. The proposed method may be beneficial to several art-related tasks, particularly visual link retrieval and historical knowledge discovery in painting datasets.

Sparse-Dense Subspace Clustering

Shuai Yang, Wenqi Zhu, Yuesheng Zhu

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

Abstract Slides Poster Similar

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.

Exploiting Non-Linear Redundancy for Neural Model Compression

Muhammad Ahmed Shah, Raphael Olivier, Bhiksha Raj

Auto-TLDR; Compressing Deep Neural Networks with Linear Dependency

Abstract Slides Poster Similar

Deploying deep learning models with millions, even billions, of parameters is challenging given real world memory, power and compute constraints. In an effort to make these models more practical, in this paper, we propose a novel model compression approach that exploits linear dependence between the activations in a layer to eliminate entire structural units (neurons/convolutional filters). Our approach also adjusts the weights of the layer in a manner that is provably lossless while training if the removed neuron was perfectly predictable. We combine this approach with an annealing algorithm that may be applied during training, or even on a trained model, and demonstrate, using popular datasets, that our technique can reduce the parameters of VGG and AlexNet by more than 97\% on \cifar, 85\% on \caltech, and 19\% on ImageNet at less than 2\% loss in accuracy. Furthermore, we provide theoretical results showing that in overparametrized, locally linear (ReLU) neural networks where redundant features exist, and with correct hyperparameter selection, our method is indeed able to capture and suppress those dependencies.

Speeding-Up Pruning for Artificial Neural Networks: Introducing Accelerated Iterative Magnitude Pruning

Marco Zullich, Eric Medvet, Felice Andrea Pellegrino, Alessio Ansuini

Auto-TLDR; Iterative Pruning of Artificial Neural Networks with Overparametrization

Abstract Slides Poster Similar

In recent years, Artificial Neural Networks (ANNs) pruning has become the focal point of many researches, due to the extreme overparametrization of such models. This has urged the scientific world to investigate methods for the simplification of the structure of weights in ANNs, mainly in an effort to reduce time for both training and inference. Frankle and Carbin and later Renda, Frankle, and Carbin introduced and refined an iterative pruning method which is able to effectively prune the network of a great portion of its parameters with little to no loss in performance. On the downside, this method requires a large amount of time for its application, since, for each iteration, the network has to be trained for (almost) the same amount of epochs of the unpruned network. In this work, we show that, for a limited setting, if targeting high overall sparsity rates, this time can be effectively reduced for each iteration, save for the last one, by more than 50%, while yielding a final product (i.e., final pruned network) whose performance is comparable to the ANN obtained using the existing method.

On Learning Random Forests for Random Forest Clustering

Manuele Bicego, Francisco Escolano

Auto-TLDR; Learning Random Forests for Clustering

Abstract 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.

Separation of Aleatoric and Epistemic Uncertainty in Deterministic Deep Neural Networks

Denis Huseljic, Bernhard Sick, Marek Herde, Daniel Kottke

Auto-TLDR; AE-DNN: Modeling Uncertainty in Deep Neural Networks

Abstract Slides Poster Similar

Despite the success of deep neural networks (DNN) in many applications, their ability to model uncertainty is still significantly limited. For example, in safety-critical applications such as autonomous driving, it is crucial to obtain a prediction that reflects different types of uncertainty to address life-threatening situations appropriately. In such cases, it is essential to be aware of the risk (i.e., aleatoric uncertainty) and the reliability (i.e., epistemic uncertainty) that comes with a prediction. We present AE-DNN, a model allowing the separation of aleatoric and epistemic uncertainty while maintaining a proper generalization capability. AE-DNN is based on deterministic DNN, which can determine the respective uncertainty measures in a single forward pass. In analyses with synthetic and image data, we show that our method improves the modeling of epistemic uncertainty while providing an intuitively understandable separation of risk and reliability.

Trainable Spectrally Initializable Matrix Transformations in Convolutional Neural Networks

Michele Alberti, Angela Botros, Schuetz Narayan, Rolf Ingold, Marcus Liwicki, Mathias Seuret

Auto-TLDR; Trainable and Spectrally Initializable Matrix Transformations for Neural Networks

Abstract Slides Poster Similar

In this work, we introduce a new architectural component to Neural Networks (NN), i.e., trainable and spectrally initializable matrix transformations on feature maps. While previous literature has already demonstrated the possibility of adding static spectral transformations as feature processors, our focus is on more general trainable transforms. We study the transforms in various architectural configurations on four datasets of different nature: from medical (ColorectalHist, HAM10000) and natural (Flowers) images to historical documents (CB55). With rigorous experiments that control for the number of parameters and randomness, we show that networks utilizing the introduced matrix transformations outperform vanilla neural networks. The observed accuracy increases appreciably across all datasets. In addition, we show that the benefit of spectral initialization leads to significantly faster convergence, as opposed to randomly initialized matrix transformations. The transformations are implemented as auto-differentiable PyTorch modules that can be incorporated into any neural network architecture. The entire code base is open-source.

An Invariance-Guided Stability Criterion for Time Series Clustering Validation

Florent Forest, Alex Mourer, Mustapha Lebbah, Hanane Azzag, Jérôme Lacaille

Auto-TLDR; An invariance-guided method for clustering model selection in time series data

Abstract Slides Poster Similar

Time series clustering is a challenging task due to the specificities of this type of data. Temporal correlation and invariance to transformations such as shifting, warping or noise prevent the use of standard data mining methods. Time series clustering has been mostly studied under the angle of finding efficient algorithms and distance metrics adapted to the specific nature of time series data. Much less attention has been devoted to the general problem of model selection. Clustering stability has emerged as a universal and model-agnostic principle for clustering model selection. This principle can be stated as follows: an algorithm should find a structure in the data that is resilient to perturbation by sampling or noise. We propose to apply stability analysis to time series by leveraging prior knowledge on the nature and invariances of the data. These invariances determine the perturbation process used to assess stability. Based on a recently introduced criterion combining between-cluster and within-cluster stability, we propose an invariance-guided method for model selection, applicable to a wide range of clustering algorithms. Experiments conducted on artificial and benchmark data sets demonstrate the ability of our criterion to discover structure and select the correct number of clusters, whenever data invariances are known beforehand.

Label Self-Adaption Hashing for Image Retrieval

Jianglin Lu, Zhihui Lai, Hailing Wang, Jie Zhou

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

Abstract 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.

Contextual Classification Using Self-Supervised Auxiliary Models for Deep Neural Networks

Sebastian Palacio, Philipp Engler, Jörn Hees, Andreas Dengel

Auto-TLDR; Self-Supervised Autogenous Learning for Deep Neural Networks

Abstract Slides Poster Similar

Classification problems solved with deep neural networks (DNNs) typically rely on a closed world paradigm, and optimize over a single objective (e.g., minimization of the cross- entropy loss). This setup dismisses all kinds of supporting signals that can be used to reinforce the existence or absence of particular patterns. The increasing need for models that are interpretable by design makes the inclusion of said contextual signals a crucial necessity. To this end, we introduce the notion of Self-Supervised Autogenous Learning (SSAL). A SSAL objective is realized through one or more additional targets that are derived from the original supervised classification task, following architectural principles found in multi-task learning. SSAL branches impose low-level priors into the optimization process (e.g., grouping). The ability of using SSAL branches during inference, allow models to converge faster, focusing on a richer set of class-relevant features. We equip state-of-the-art DNNs with SSAL objectives and report consistent improvements for all of them on CIFAR100 and Imagenet. We show that SSAL models outperform similar state-of-the-art methods focused on contextual loss functions, auxiliary branches and hierarchical priors.

Emerging Relation Network and Task Embedding for Multi-Task Regression Problems

Schreiber Jens, Bernhard Sick

Auto-TLDR; A Comparative Study of Multi-Task Learning for Non-linear Time Series Problems

Abstract Slides Poster Similar

Multi-Task learning (MTL) provides state-of-the-art results in many applications of computer vision and natural language processing. In contrast to single-task learning (STL), MTL allows for leveraging knowledge between related tasks improving prediction results on all tasks. However, there is a limited number of comparative studies applied to MTL architectures for regression and time series problems taking recent advances of MTL into account. An intriguing, non-linear time-series problem are day ahead forecasts of the expected power generation for renewable power plants. Therefore, the main contribution of this article is a comparative study of the following recent and relevant MTL architectures: Hard-parameter sharing, cross-stitch network, and sluice network (SN). They are compared to a multi-layer peceptron (MLP) model of similar size in an STL setting. As a additional contribution, we provide a simple, yet practical approach to model task specific information through an embedding layer in an MLP, referred to as task embedding. Further, we contribute a new MTL architecture named emerging relation network (ERN), which can be considered as an extension of the SN. For a solar power dataset, the task embedding achieves the best mean improvement with 8.2%. For two wind and one additional solar dataset, the ERN is the best MTL architecture with improvements up to 11.3%.

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

Yuanhang Ren, Ye Du

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

Abstract 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.

How to Define a Rejection Class Based on Model Learning?

Sarah Laroui, Xavier Descombes, Aurelia Vernay, Florent Villiers, Francois Villalba, Eric Debreuve

Auto-TLDR; An innovative learning strategy for supervised classification that is able, by design, to reject a sample as not belonging to any of the known classes

Abstract Slides Poster Similar

In supervised classification, the learning process typically trains a classifier to optimize the accuracy of classifying data into the classes that appear in the learning set, and only them. While this framework fits many use cases, there are situations where the learning process is knowingly performed using a learning set that only represents the data that have been observed so far among a virtually unconstrained variety of possible samples. It is then crucial to define a classifier which has the ability to reject a sample, i.e., to classify it into a rejection class that has not been yet defined. Although obvious solutions can add this ability a posteriori to a classifier that has been learned classically, a better approach seems to directly account for this requirement in the classifier design. In this paper, we propose an innovative learning strategy for supervised classification that is able, by design, to reject a sample as not belonging to any of the known classes. For that, we rely on modeling each class as the combination of a probability density function (PDF) and a threshold that is computed with respect to the other classes. Several alternatives are proposed and compared in this framework. A comparison with straightforward approaches is also provided.

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

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

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

Abstract 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.