Current deep learning based video classification architectures are typically trained end-to-end on large volumes of data and require extensive computational resources. This paper aims to exploit audio-visual information in video classification with a 1 frame per second sampling rate. We propose Temporal Bilinear Encoding Networks (TBEN) for encoding both audio and visual long range temporal information using bilinear pooling and demonstrate bilinear pooling is better than average pooling on the temporal dimension for videos with low sampling rate. We also embed the label hierarchy in TBEN to further improve the robustness of the classifier. Experiments on the FGA240 fine-grained classification dataset using TBEN achieve a new state-of-the-art (hit@1=47.95%). We also exploit the possibility of incorporating TBEN with multiple decoupled modalities like visual semantic and motion features: experiments on UCF101 sampled at 1 FPS achieve close to state-of-the-art accuracy (hit@1=91.03%) while requiring significantly less computational resources than competing approaches for both training and prediction.
Spectral graph convolutional networks are generalizations of standard convolutional networks for graph-structured data using the Laplacian operator. A common misconception is the instability of spectral filters, i.e. the impossibility to transfer spectral filters between graphs of variable size and topology. This misbelief has limited the development of spectral networks for multi-graph tasks in favor of spatial graph networks. However, recent works have proved the stability of spectral filters under graph perturbation. Our work complements and emphasizes further the high quality of spectral transferability by benchmarking spectral graph networks on tasks involving graphs of different size and connectivity. Numerical experiments exhibit favorable performance on graph regression, graph classification, and node classification problems on two graph benchmarks. The implementation of our experiments is available on GitHub for reproducibility.
Standard approaches to group-based notions of fairness, such as \emph{parity} and \emph{equalized odds}, try to equalize absolute measures of performance across known groups (based on race, gender, etc.). Consequently, a group that is inherently harder to classify may hold back the performance on other groups; and no guarantees can be provided for unforeseen groups. Instead, we propose a fairness notion whose guarantee, on each group $g$ in a class $\mathcal{G}$, is relative to the performance of the best classifier on $g$. We apply this notion to broad classes of groups, in particular, where (a) $\mathcal{G}$ consists of all possible groups (subsets) in the data, and (b) $\mathcal{G}$ is more streamlined. For the first setting, which is akin to groups being completely unknown, we devise the {\sc PF} (Proportional Fairness) classifier, which guarantees, on any possible group $g$, an accuracy that is proportional to that of the optimal classifier for $g$, scaled by the relative size of $g$ in the data set. Due to including all possible groups, some of which could be too complex to be relevant, the worst-case theoretical guarantees here have to be proportionally weaker for smaller subsets. For the second setting, we devise the {\sc BeFair} (Best-effort Fair) framework which seeks an accuracy, on every $g \in \mathcal{G}$, which approximates that of the optimal classifier on $g$, independent of the size of $g$. Aiming for such a guarantee results in a non-convex problem, and we design novel techniques to get around this difficulty when $\mathcal{G}$ is the set of linear hypotheses. We test our algorithms on real-world data sets, and present interesting comparative insights on their performance.
Machine learning techniques can be useful in applications such as credit approval and college admission. However, to be classified more favorably in such contexts, an agent may decide to strategically withhold some of her features, such as bad test scores. This is a missing data problem with a twist: which data is missing {\em depends on the chosen classifier}, because the specific classifier is what may create the incentive to withhold certain feature values. We address the problem of training classifiers that are robust to this behavior. We design three classification methods: {\sc Mincut}, {\sc Hill-Climbing} ({\sc HC}) and Incentive-Compatible Logistic Regression ({\sc IC-LR}). We show that {\sc Mincut} is optimal when the true distribution of data is fully known. However, it can produce complex decision boundaries, and hence be prone to overfitting in some cases. Based on a characterization of truthful classifiers (i.e., those that give no incentive to strategically hide features), we devise a simpler alternative called {\sc HC} which consists of a hierarchical ensemble of out-of-the-box classifiers, trained using a specialized hill-climbing procedure which we show to be convergent. For several reasons, {\sc Mincut} and {\sc HC} are not effective in utilizing a large number of complementarily informative features. To this end, we present {\sc IC-LR}, a modification of Logistic Regression that removes the incentive to strategically drop features. We also show that our algorithms perform well in experiments on real-world data sets, and present insights into their relative performance in different settings.
Object point cloud classification has drawn great research attention since the release of benchmarking datasets, such as the ModelNet and the ShapeNet. These benchmarks assume point clouds covering complete surfaces of object instances, for which plenty of high-performing methods have been developed. However, their settings deviate from those often met in practice, where, due to (self-)occlusion, a point cloud covering partial surface of an object is captured from an arbitrary view. We show in this paper that performance of existing point cloud classifiers drops drastically under the considered single-view, partial setting; the phenomenon is consistent with the observation that semantic category of a partial object surface is less ambiguous only when its distribution on the whole surface is clearly specified. To this end, we argue for a single-view, partial setting where supervised learning of object pose estimation should be accompanied with classification. Technically, we propose a baseline method of Pose-Accompanied Point cloud classification Network (PAPNet); built upon SE(3)-equivariant convolutions, the PAPNet learns intermediate pose transformations for equivariant features defined on vector fields, which makes the subsequent classification easier (ideally) in the category-level, canonical pose. By adapting existing ModelNet40 and ScanNet datasets to the single-view, partial setting, experiment results can verify the necessity of object pose estimation and superiority of our PAPNet to existing classifiers.
Spurious correlations threaten the validity of statistical classifiers. While model accuracy may appear high when the test data is from the same distribution as the training data, it can quickly degrade when the test distribution changes. For example, it has been shown that classifiers perform poorly when humans make minor modifications to change the label of an example. One solution to increase model reliability and generalizability is to identify causal associations between features and classes. In this paper, we propose to train a robust text classifier by augmenting the training data with automatically generated counterfactual data. We first identify likely causal features using a statistical matching approach. Next, we generate counterfactual samples for the original training data by substituting causal features with their antonyms and then assigning opposite labels to the counterfactual samples. Finally, we combine the original data and counterfactual data to train a robust classifier. Experiments on two classification tasks show that a traditional classifier trained on the original data does very poorly on human-generated counterfactual samples (e.g., 10%-37% drop in accuracy). However, the classifier trained on the combined data is more robust and performs well on both the original test data and the counterfactual test data (e.g., 12%-25% increase in accuracy compared with the traditional classifier). Detailed analysis shows that the robust classifier makes meaningful and trustworthy predictions by emphasizing causal features and de-emphasizing non-causal features.
Electroencephalography is frequently used for diagnostic evaluation of various brain-related disorders due to its excellent resolution, non-invasive nature and low cost. However, manual analysis of EEG signals could be strenuous and a time-consuming process for experts. It requires long training time for physicians to develop expertise in it and additionally experts have low inter-rater agreement (IRA) among themselves. Therefore, many Computer Aided Diagnostic (CAD) based studies have considered the automation of interpreting EEG signals to alleviate the workload and support the final diagnosis. In this paper, we present an automatic binary classification framework for brain signals in multichannel EEG recordings. We propose to use Wavelet Packet Decomposition (WPD) techniques to decompose the EEG signals into frequency sub-bands and extract a set of statistical features from each of the selected coefficients. Moreover, we propose a novel method to reduce the dimension of the feature space without compromising the quality of the extracted features. The extracted features are classified using different Gradient Boosting Decision Tree (GBDT) based classification frameworks, which are CatBoost, XGBoost and LightGBM. We used Temple University Hospital EEG Abnormal Corpus V2.0.0 to test our proposed technique. We found that CatBoost classifier achieves the binary classification accuracy of 87.68%, and outperforms state-of-the-art techniques on the same dataset by more than 1% in accuracy and more than 3% in sensitivity. The obtained results in this research provide important insights into the usefulness of WPD feature extraction and GBDT classifiers for EEG classification.
A heterogeneous information network (HIN) has as vertices objects of different types and as edges the relations between objects, which are also of various types. We study the problem of classifying objects in HINs. Most existing methods perform poorly when given scarce labeled objects as training sets, and methods that improve classification accuracy under such scenarios are often computationally expensive. To address these problems, we propose ConCH, a graph neural network model. ConCH formulates the classification problem as a multi-task learning problem that combines semi-supervised learning with self-supervised learning to learn from both labeled and unlabeled data. ConCH employs meta-paths, which are sequences of object types that capture semantic relationships between objects. ConCH co-derives object embeddings and context embeddings via graph convolution. It also uses the attention mechanism to fuse such embeddings. We conduct extensive experiments to evaluate the performance of ConCH against other 15 classification methods. Our results show that ConCH is an effective and efficient method for HIN classification.
Numerical experiments demonstrate that deep neural network classifiers progressively separate class distributions around their mean, achieving linear separability on the training set, and increasing the Fisher discriminant ratio. We explain this mechanism with two types of operators. We prove that a rectifier without biases applied to sign-invariant tight frames can separate class means and increase Fisher ratios. On the opposite, a soft-thresholding on tight frames can reduce within-class variabilities while preserving class means. Variance reduction bounds are proved for Gaussian mixture models. For image classification, we show that separation of class means can be achieved with rectified wavelet tight frames that are not learned. It defines a scattering transform. Learning $1 \times 1$ convolutional tight frames along scattering channels and applying a soft-thresholding reduces within-class variabilities. The resulting scattering network reaches the classification accuracy of ResNet-18 on CIFAR-10 and ImageNet, with fewer layers and no learned biases.
Open intent classification is a challenging task in dialogue systems. On the one hand, it should ensure the quality of known intent identification. On the other hand, it needs to detect the open (unknown) intent without prior knowledge. Current models are limited in finding the appropriate decision boundary to balance the performances of both known intents and the open intent. In this paper, we propose a post-processing method to learn the adaptive decision boundary (ADB) for open intent classification. We first utilize the labeled known intent samples to pre-train the model. Then, we automatically learn the adaptive spherical decision boundary for each known class with the aid of well-trained features. Specifically, we propose a new loss function to balance both the empirical risk and the open space risk. Our method does not need open intent samples and is free from modifying the model architecture. Moreover, our approach is surprisingly insensitive with less labeled data and fewer known intents. Extensive experiments on three benchmark datasets show that our method yields significant improvements compared with the state-of-the-art methods. The codes are released at https://github.com/thuiar/Adaptive-Decision-Boundary.
Japanese patents are assigned a patent classification code, FI (File Index), that is unique to Japan. FI is a subdivision of the IPC, an international patent classification code, that is related to Japanese technology. FIs are revised to keep up with technological developments. These revisions have already established more than 30,000 new FIs since 2006. However, these revisions require a lot of time and workload. Moreover, these revisions are not automated and are thus inefficient. Therefore, using machine learning to assist in the revision of patent classification codes (FI) will lead to improved accuracy and efficiency. This study analyzes patent documents from this new perspective of assisting in the revision of patent classification codes with machine learning. To analyze time-series changes in patents, we used the dynamic topic model (DTM), which is an extension of the latent Dirichlet allocation (LDA). Also, unlike English, the Japanese language requires morphological analysis. Patents contain many technical words that are not used in everyday life, so morphological analysis using a common dictionary is not sufficient. Therefore, we used a technique for extracting technical terms from text. After extracting technical terms, we applied them to DTM. In this study, we determined the technological progress of the lighting class F21 for 14 years and compared it with the actual revision of patent classification codes. In other words, we extracted technical terms from Japanese patents and applied DTM to determine the progress of Japanese technology. Then, we analyzed the results from the new perspective of revising patent classification codes with machine learning. As a result, it was found that those whose topics were on the rise were judged to be new technologies.
Deep Learning approaches have brought solutions, with impressive performance, to general classification problems where wealthy of annotated data are provided for training. In contrast, less progress has been made in continual learning of a set of non-stationary classes, mainly when applied to unsupervised problems with streaming data. Here, we propose a novel incremental learning approach which combines a deep features encoder with an Open-Set Dynamic Ensembles of SVM, to tackle the problem of identifying individuals of interest (IoI) from streaming face data. From a simple weak classifier trained on a few video-frames, our method can use unsupervised operational data to enhance recognition. Our approach adapts to new patterns avoiding catastrophic forgetting and partially heals itself from miss-adaptation. Besides, to better comply with real world conditions, the system was designed to operate in an open-set setting. Results show a benefit of up to 15% F1-score increase respect to non-adaptive state-of-the-art methods.
We present a machine learning (ML) pipeline to identify star clusters in the multi{color images of nearby galaxies, from observations obtained with the Hubble Space Telescope as part of the Treasury Project LEGUS (Legacy ExtraGalactic Ultraviolet Survey). StarcNet (STAR Cluster classification NETwork) is a multi-scale convolutional neural network (CNN) which achieves an accuracy of 68.6% (4 classes)/86.0% (2 classes: cluster/non-cluster) for star cluster classification in the images of the LEGUS galaxies, nearly matching human expert performance. We test the performance of StarcNet by applying pre-trained CNN model to galaxies not included in the training set, finding accuracies similar to the reference one. We test the effect of StarcNet predictions on the inferred cluster properties by comparing multi-color luminosity functions and mass-age plots from catalogs produced by StarcNet and by human-labeling; distributions in luminosity, color, and physical characteristics of star clusters are similar for the human and ML classified samples. There are two advantages to the ML approach: (1) reproducibility of the classifications: the ML algorithm's biases are fixed and can be measured for subsequent analysis; and (2) speed of classification: the algorithm requires minutes for tasks that humans require weeks to months to perform. By achieving comparable accuracy to human classifiers, StarcNet will enable extending classifications to a larger number of candidate samples than currently available, thus increasing significantly the statistics for cluster studies.
We propose a kernelized classification layer for deep networks. Although conventional deep networks introduce an abundance of nonlinearity for representation (feature) learning, they almost universally use a linear classifier on the learned feature vectors. We advocate a nonlinear classification layer by using the kernel trick on the softmax cross-entropy loss function during training and the scorer function during testing. However, the choice of the kernel remains a challenge. To tackle this, we theoretically show the possibility of optimizing over all possible positive definite kernels applicable to our problem setting. This theory is then used to device a new kernelized classification layer that learns the optimal kernel function for a given problem automatically within the deep network itself. We show the usefulness of the proposed nonlinear classification layer on several datasets and tasks.
In this paper, we provide a guideline for using the Neural Network Dependability Kit (NNDK) during the development process of NN models, and show how the algorithm is applied in two image classification use cases. The case studies demonstrate the usage of the dependability kit to obtain insights about the NN model and how they informed the development process of the neural network model. After interpreting neural networks via the different metrics available in the NNDK, the developers were able to increase the NNs' accuracy, trust the developed networks, and make them more robust. In addition, we obtained a novel application-oriented technique to provide supporting evidence for an NN's classification result to the user. In the medical image classification use case, it was used to retrieve case images from the training dataset that were similar to the current patient's image and could therefore act as a support for the NN model's decision and aid doctors in interpreting the results.
3D Convolutional Neural Network (3D CNN) captures spatial and temporal information on 3D data such as video sequences. However, due to the convolution and pooling mechanism, the information loss seems unavoidable. To improve the visual explanations and classification in 3D CNN, we propose two approaches; i) aggregate layer-wise global to local (global-local) discrete gradients using trained 3DResNext network, and ii) implement attention gating network to improve the accuracy of the action recognition. The proposed approach intends to show the usefulness of every layer termed as global-local attention in 3D CNN via visual attribution, weakly-supervised action localization, and action recognition. Firstly, the 3DResNext is trained and applied for action classification using backpropagation concerning the maximum predicted class. The gradients and activations of every layer are then up-sampled. Later, aggregation is used to produce more nuanced attention, which points out the most critical part of the predicted class's input videos. We use contour thresholding of final attention for final localization. We evaluate spatial and temporal action localization in trimmed videos using fine-grained visual explanation via 3DCam. Experimental results show that the proposed approach produces informative visual explanations and discriminative attention. Furthermore, the action recognition via attention gating on each layer produces better classification results than the baseline model.
Using evolutionary computation algorithms to solve multiple tasks with knowledge sharing is a promising approach. Image feature learning can be considered as a multitask problem because different tasks may have a similar feature space. Genetic programming (GP) has been successfully applied to image feature learning for classification. However, most of the existing GP methods solve one task, independently, using sufficient training data. No multitask GP method has been developed for image feature learning. Therefore, this paper develops a multitask GP approach to image feature learning for classification with limited training data. Owing to the flexible representation of GP, a new knowledge sharing mechanism based on a new individual representation is developed to allow GP to automatically learn what to share across two tasks and to improve its learning performance. The shared knowledge is encoded as a common tree, which can represent the common/general features of two tasks. With the new individual representation, each task is solved using the features extracted from a common tree and a task-specific tree representing task-specific features. To learn the best common and task-specific trees, a new evolutionary process and new fitness functions are developed. The performance of the proposed approach is examined on six multitask problems of 12 image classification datasets with limited training data and compared with three GP and 14 non-GP-based competitive methods. Experimental results show that the new approach outperforms these compared methods in almost all the comparisons. Further analysis reveals that the new approach learns simple yet effective common trees with high effectiveness and transferability.
This work studies constrained stochastic optimization problems where the objective and constraint functions are convex and expressed as compositions of stochastic functions. The problem arises in the context of fair classification, fair regression, and the design of queuing systems. Of particular interest is the large-scale setting where an oracle provides the stochastic gradients of the constituent functions, and the goal is to solve the problem with a minimal number of calls to the oracle. Owing to the compositional form, the stochastic gradients provided by the oracle do not yield unbiased estimates of the objective or constraint gradients. Instead, we construct approximate gradients by tracking the inner function evaluations, resulting in a quasi-gradient saddle point algorithm. We prove that the proposed algorithm is guaranteed to find the optimal and feasible solution almost surely. We further establish that the proposed algorithm requires $\mathcal{O}(1/\epsilon^4)$ data samples in order to obtain an $\epsilon$-approximate optimal point while also ensuring zero constraint violation. The result matches the sample complexity of the stochastic compositional gradient descent method for unconstrained problems and improves upon the best-known sample complexity results for the constrained settings. The efficacy of the proposed algorithm is tested on both fair classification and fair regression problems. The numerical results show that the proposed algorithm outperforms the state-of-the-art algorithms in terms of the convergence rate.
We introduce a new dataset, MELINDA, for Multimodal biomEdicaL experImeNt methoD clAssification. The dataset is collected in a fully automated distant supervision manner, where the labels are obtained from an existing curated database, and the actual contents are extracted from papers associated with each of the records in the database. We benchmark various state-of-the-art NLP and computer vision models, including unimodal models which only take either caption texts or images as inputs, and multimodal models. Extensive experiments and analysis show that multimodal models, despite outperforming unimodal ones, still need improvements especially on a less-supervised way of grounding visual concepts with languages, and better transferability to low resource domains. We release our dataset and the benchmarks to facilitate future research in multimodal learning, especially to motivate targeted improvements for applications in scientific domains.
Self-attention techniques, and specifically Transformers, are dominating the field of text processing and are becoming increasingly popular in computer vision classification tasks. In order to visualize the parts of the image that led to a certain classification, existing methods either rely on the obtained attention maps or employ heuristic propagation along the attention graph. In this work, we propose a novel way to compute relevancy for Transformer networks. The method assigns local relevance based on the Deep Taylor Decomposition principle and then propagates these relevancy scores through the layers. This propagation involves attention layers and skip connections, which challenge existing methods. Our solution is based on a specific formulation that is shown to maintain the total relevancy across layers. We benchmark our method on very recent visual Transformer networks, as well as on a text classification problem, and demonstrate a clear advantage over the existing explainability methods.