Outcome labeling ambiguity and subjectivity are ubiquitous in real-world datasets. While practitioners commonly combine ambiguous outcome labels for all data points (instances) in an ad hoc way to improve the accuracy of multi-class classification, there lacks a principled approach to guide the label combination for all data points by any optimality criterion. To address this problem, we propose the information-theoretic classification accuracy (ITCA), a criterion that balances the trade-off between prediction accuracy (how well do predicted labels agree with actual labels) and classification resolution (how many labels are predictable), to guide practitioners on how to combine ambiguous outcome labels. To find the optimal label combination indicated by ITCA, we propose two search strategies: greedy search and breadth-first search. Notably, ITCA and the two search strategies are adaptive to all machine-learning classification algorithms. Coupled with a classification algorithm and a search strategy, ITCA has two uses: improving prediction accuracy and identifying ambiguous labels. We first verify that ITCA achieves high accuracy with both search strategies in finding the correct label combinations on synthetic and real data. Then we demonstrate the effectiveness of ITCA in diverse applications including medical prognosis, cancer survival prediction, user demographics prediction, and cell type classification. We also provide theoretical insights into ITCA by studying the oracle and the linear discriminant analysis classification algorithms. Python package itca (available at https://github.com/JSB-UCLA/ITCA) implements ITCA and search strategies.
The predict-then-optimize framework is fundamental in practical stochastic decision-making problems: first predict unknown parameters of an optimization model, then solve the problem using the predicted values. A natural loss function in this setting is defined by measuring the decision error induced by the predicted parameters, which was named the Smart Predict-then-Optimize (SPO) loss by Elmachtoub and Grigas [arXiv:1710.08005]. Since the SPO loss is typically nonconvex and possibly discontinuous, Elmachtoub and Grigas [arXiv:1710.08005] introduced a convex surrogate, called the SPO+ loss, that importantly accounts for the underlying structure of the optimization model. In this paper, we greatly expand upon the consistency results for the SPO+ loss provided by Elmachtoub and Grigas [arXiv:1710.08005]. We develop risk bounds and uniform calibration results for the SPO+ loss relative to the SPO loss, which provide a quantitative way to transfer the excess surrogate risk to excess true risk. By combining our risk bounds with generalization bounds, we show that the empirical minimizer of the SPO+ loss achieves low excess true risk with high probability. We first demonstrate these results in the case when the feasible region of the underlying optimization problem is a polyhedron, and then we show that the results can be strengthened substantially when the feasible region is a level set of a strongly convex function. We perform experiments to empirically demonstrate the strength of the SPO+ surrogate, as compared to standard $\ell_1$ and squared $\ell_2$ prediction error losses, on portfolio allocation and cost-sensitive multi-class classification problems.
Semi-supervised learning (SSL) has proven to be effective at leveraging large-scale unlabeled data to mitigate the dependency on labeled data in order to learn better models for visual recognition and classification tasks. However, recent SSL methods rely on unlabeled image data at a scale of billions to work well. This becomes infeasible for tasks with relatively fewer unlabeled data in terms of runtime, memory and data acquisition. To address this issue, we propose noisy semi-supervised transfer learning, an efficient SSL approach that integrates transfer learning and self-training with noisy student into a single framework, which is tailored for tasks that can leverage unlabeled image data on a scale of thousands. We evaluate our method on both binary and multi-class classification tasks, where the objective is to identify whether an image displays people practicing sports or the type of sport, as well as to identify the pose from a pool of popular yoga poses. Extensive experiments and ablation studies demonstrate that by leveraging unlabeled data, our proposed framework significantly improves visual classification, especially in multi-class classification settings compared to state-of-the-art methods. Moreover, incorporating transfer learning not only improves classification performance, but also requires 6x less compute time and 5x less memory. We also show that our method boosts robustness of visual classification models, even without specifically optimizing for adversarial robustness.
Multi-label learning is an emerging extension of the multi-class classification where an image contains multiple labels. Not only acquiring a clean and fully labeled dataset in multi-label learning is extremely expensive, but also many of the actual labels are corrupted or missing due to the automated or non-expert annotation techniques. Noisy label data decrease the prediction performance drastically. In this paper, we propose a novel Gold Asymmetric Loss Correction with Single-Label Regulators (GALC-SLR) that operates robust against noisy labels. GALC-SLR estimates the noise confusion matrix using single-label samples, then constructs an asymmetric loss correction via estimated confusion matrix to avoid overfitting to the noisy labels. Empirical results show that our method outperforms the state-of-the-art original asymmetric loss multi-label classifier under all corruption levels, showing mean average precision improvement up to 28.67% on a real world dataset of MS-COCO, yielding a better generalization of the unseen data and increased prediction performance.
State-of-the-art deep face recognition methods are mostly trained with a softmax-based multi-class classification framework. Despite being popular and effective, these methods still have a few shortcomings that limit empirical performance. In this paper, we start by identifying the discrepancy between training and evaluation in the existing multi-class classification framework and then discuss the potential limitations caused by the "competitive" nature of softmax normalization. Motivated by these limitations, we propose a novel binary classification training framework, termed SphereFace2. In contrast to existing methods, SphereFace2 circumvents the softmax normalization, as well as the corresponding closed-set assumption. This effectively bridges the gap between training and evaluation, enabling the representations to be improved individually by each binary classification task. Besides designing a specific well-performing loss function, we summarize a few general principles for this "one-vs-all" binary classification framework so that it can outperform current competitive methods. Our experiments on popular benchmarks demonstrate that SphereFace2 can consistently outperform state-of-the-art deep face recognition methods. The code has been made publicly available.
In this paper, Bayesian based aggregation of decision trees in an ensemble (decision forest) is investigated. The focus is laid on multi-class classification with number of samples significantly skewed toward one of the classes. The algorithm leverages out-of-bag datasets to estimate prediction errors of individual trees, which are then used in accordance with the Bayes rule to refine the decision of the ensemble. The algorithm takes prevalence of individual classes into account and does not require setting of any additional parameters related to class weights or decision-score thresholds. Evaluation is based on publicly available datasets as well as on an proprietary dataset comprising network traffic telemetry from hundreds of enterprise networks with over a million of users overall. The aim is to increase the detection capabilities of an operating malware detection system. While we were able to keep precision of the system higher than 94\%, that is only 6 out of 100 detections shown to the network administrator are false alarms, we were able to achieve increase of approximately 7\% in the number of detections. The algorithm effectively handles large amounts of data, and can be used in conjunction with most of the state-of-the-art algorithms used to train decision forests.
Online educational platforms organize academic questions based on a hierarchical learning taxonomy (subject-chapter-topic). Automatically tagging new questions with existing taxonomy will help organize these questions into different classes of hierarchical taxonomy so that they can be searched based on the facets like chapter. This task can be formulated as a flat multi-class classification problem. Usually, flat classification based methods ignore the semantic relatedness between the terms in the hierarchical taxonomy and the questions. Some traditional methods also suffer from the class imbalance issues as they consider only the leaf nodes ignoring the hierarchy. Hence, we formulate the problem as a similarity-based retrieval task where we optimize the semantic relatedness between the taxonomy and the questions. We demonstrate that our method helps to handle the unseen labels and hence can be used for taxonomy tagging in the wild. In this method, we augment the question with its corresponding answer to capture more semantic information and then align the question-answer pair's contextualized embedding with the corresponding label (taxonomy) vector representations. The representations are aligned by fine-tuning a transformer based model with a loss function that is a combination of the cosine similarity and hinge rank loss. The loss function maximizes the similarity between the question-answer pair and the correct label representations and minimizes the similarity to unrelated labels. Finally, we perform experiments on two real-world datasets. We show that the proposed learning method outperforms representations learned using the multi-class classification method and other state of the art methods by 6% as measured by Recall@k. We also demonstrate the performance of the proposed method on unseen but related learning content like the learning objectives without re-training the network.
The newly emerged transformer technology has a tremendous impact on NLP research. In the general English domain, transformer-based models have achieved state-of-the-art performances on various NLP benchmarks. In the clinical domain, researchers also have investigated transformer models for clinical applications. The goal of this study is to systematically explore three widely used transformer-based models (i.e., BERT, RoBERTa, and XLNet) for clinical relation extraction and develop an open-source package with clinical pre-trained transformer-based models to facilitate information extraction in the clinical domain. We developed a series of clinical RE models based on three transformer architectures, namely BERT, RoBERTa, and XLNet. We evaluated these models using 2 publicly available datasets from 2018 MADE1.0 and 2018 n2c2 challenges. We compared two classification strategies (binary vs. multi-class classification) and investigated two approaches to generate candidate relations in different experimental settings. In this study, we compared three transformer-based (BERT, RoBERTa, and XLNet) models for relation extraction. We demonstrated that the RoBERTa-clinical RE model achieved the best performance on the 2018 MADE1.0 dataset with an F1-score of 0.8958. On the 2018 n2c2 dataset, the XLNet-clinical model achieved the best F1-score of 0.9610. Our results indicated that the binary classification strategy consistently outperformed the multi-class classification strategy for clinical relation extraction. Our methods and models are publicly available at https://github.com/uf-hobi-informatics-lab/ClinicalTransformerRelationExtraction. We believe this work will improve current practice on clinical relation extraction and other related NLP tasks in the biomedical domain.
Modern keyboards allow a musician to play multiple instruments at the same time by assigning zones -- fixed pitch ranges of the keyboard -- to different instruments. In this paper, we aim to further extend this idea and examine the feasibility of automatic instrumentation -- dynamically assigning instruments to notes in solo music during performance. In addition to the online, real-time-capable setting for performative use cases, automatic instrumentation can also find applications in assistive composing tools in an offline setting. Due to the lack of paired data of original solo music and their full arrangements, we approach automatic instrumentation by learning to separate parts (e.g., voices, instruments and tracks) from their mixture in symbolic multitrack music, assuming that the mixture is to be played on a keyboard. We frame the task of part separation as a sequential multi-class classification problem and adopt machine learning to map sequences of notes into sequences of part labels. To examine the effectiveness of our proposed models, we conduct a comprehensive empirical evaluation over four diverse datasets of different genres and ensembles -- Bach chorales, string quartets, game music and pop music. Our experiments show that the proposed models outperform various baselines. We also demonstrate the potential for our proposed models to produce alternative convincing instrumentations for an existing arrangement by separating its mixture into parts. All source code and audio samples can be found at https://salu133445.github.io/arranger/ .
Tropical cyclones are one of the most powerful and destructive natural phenomena on earth. Tropical storms and heavy rains can cause floods, which lead to human lives and economic loss. Devastating winds accompanying cyclones heavily affect not only the coastal regions, even distant areas. Our study focuses on the intensity estimation, particularly cyclone grade and maximum sustained surface wind speed (MSWS) of a tropical cyclone over the North Indian Ocean. We use various machine learning algorithms to estimate cyclone grade and MSWS. We have used the basin of origin, date, time, latitude, longitude, estimated central pressure, and pressure drop as attributes of our models. We use multi-class classification models for the categorical outcome variable, cyclone grade, and regression models for MSWS as it is a continuous variable. Using the best track data of 28 years over the North Indian Ocean, we estimate grade with an accuracy of 88% and MSWS with a root mean square error (RMSE) of 2.3. For higher grade categories (5-7), accuracy improves to an average of 98.84%. We tested our model with two recent tropical cyclones in the North Indian Ocean, Vayu and Fani. For grade, we obtained an accuracy of 93.22% and 95.23% respectively, while for MSWS, we obtained RMSE of 2.2 and 3.4 and $R^2$ of 0.99 and 0.99, respectively.
Based on PixelHop and PixelHop++, which are recently developed using the successive subspace learning (SSL) framework, we propose an enhanced solution for object classification, called E-PixelHop, in this work. E-PixelHop consists of the following steps. First, to decouple the color channels for a color image, we apply principle component analysis and project RGB three color channels onto two principle subspaces which are processed separately for classification. Second, to address the importance of multi-scale features, we conduct pixel-level classification at each hop with various receptive fields. Third, to further improve pixel-level classification accuracy, we develop a supervised label smoothing (SLS) scheme to ensure prediction consistency. Forth, pixel-level decisions from each hop and from each color subspace are fused together for image-level decision. Fifth, to resolve confusing classes for further performance boosting, we formulate E-PixelHop as a two-stage pipeline. In the first stage, multi-class classification is performed to get a soft decision for each class, where the top 2 classes with the highest probabilities are called confusing classes. Then,we conduct a binary classification in the second stage. The main contributions lie in Steps 1, 3 and 5.We use the classification of the CIFAR-10 dataset as an example to demonstrate the effectiveness of the above-mentioned key components of E-PixelHop.
Functional magnetic resonance imaging (fMRI) is a neuroimaging technique that records neural activations in the brain by capturing the blood oxygen level in different regions based on the task performed by a subject. Given fMRI data, the problem of predicting the state of cognitive fatigue in a person has not been investigated to its full extent. This paper proposes tackling this issue as a multi-class classification problem by dividing the state of cognitive fatigue into six different levels, ranging from no-fatigue to extreme fatigue conditions. We built a spatio-temporal model that uses convolutional neural networks (CNN) for spatial feature extraction and a long short-term memory (LSTM) network for temporal modeling of 4D fMRI scans. We also applied a self-supervised method called MoCo (Momentum Contrast) to pre-train our model on a public dataset BOLD5000 and fine-tuned it on our labeled dataset to predict cognitive fatigue. Our novel dataset contains fMRI scans from Traumatic Brain Injury (TBI) patients and healthy controls (HCs) while performing a series of N-back cognitive tasks. This method establishes a state-of-the-art technique to analyze cognitive fatigue from fMRI data and beats previous approaches to solve this problem.
The diagnosis of blood-based diseases often involves identifying and characterizing patient blood samples. Automated methods to detect and classify blood cell subtypes have important medical applications. Automated medical image processing and analysis offers a powerful tool for medical diagnosis. In this work we tackle the problem of white blood cell classification based on the morphological characteristics of their outer contour, color. The work we would explore a set of preprocessing and segmentation (Color-based segmentation, Morphological processing, contouring) algorithms along with a set of features extraction methods (Corner detection algorithms and Histogram of Gradients(HOG)), dimensionality reduction algorithms (Principal Component Analysis(PCA)) that are able to recognize and classify through various Unsupervised(k-nearest neighbors) and Supervised (Support Vector Machine, Decision Trees, Linear Discriminant Analysis, Quadratic Discriminant Analysis, Naive Bayes) algorithms different categories of white blood cells to Eosinophil, Lymphocyte, Monocyte, and Neutrophil. We even take a step forwards to explore various Deep Convolutional Neural network architecture (Sqeezent, MobilenetV1,MobilenetV2, InceptionNet etc.) without preprocessing/segmentation and with preprocessing. We would like to explore many algorithms to identify the robust algorithm with least time complexity and low resource requirement. The outcome of this work can be a cue to selection of algorithms as per requirement for automated blood cell classification.
Can we learn a multi-class classifier from only data of a single class? We show that without any assumptions on the loss functions, models, and optimizers, we can successfully learn a multi-class classifier from only data of a single class with a rigorous consistency guarantee when confidences (i.e., the class-posterior probabilities for all the classes) are available. Specifically, we propose an empirical risk minimization framework that is loss-/model-/optimizer-independent. Instead of constructing a boundary between the given class and other classes, our method can conduct discriminative classification between all the classes even if no data from the other classes are provided. We further theoretically and experimentally show that our method can be Bayes-consistent with a simple modification even if the provided confidences are highly noisy. Then, we provide an extension of our method for the case where data from a subset of all the classes are available. Experimental results demonstrate the effectiveness of our methods.
Mild cognitive impairment (MCI) is a high-risk dementia condition which progresses to probable Alzheimer's disease (AD) at approximately 10% to 15% per year. Characterization of group-level differences between two subtypes of MCI - stable MCI (sMCI) and progressive MCI (pMCI) is the key step to understand the mechanisms of MCI progression and enable possible delay of transition from MCI to AD. Functional connectivity (FC) is considered as a promising way to study MCI progression since which may show alterations even in preclinical stages and provide substrates for AD progression. However, the representative FC patterns during AD development for different clinical groups, especially for sMCI and pMCI, have been understudied. In this work, we integrated autoencoder and multi-class classification into a single deep model and successfully learned a set of clinical group related feature vectors. Specifically, we trained two non-linear mappings which realized the mutual transformations between original FC space and the feature space. By mapping the learned clinical group related feature vectors to the original FC space, representative FCs were constructed for each group. Moreover, based on these feature vectors, our model achieves a high classification accuracy - 68% for multi-class classification (NC vs SMC vs sMCI vs pMCI vs AD).
Magnetic Resonance Imaging (MRI) is a principal diagnostic approach used in the field of radiology to create images of the anatomical and physiological structure of patients. MRI is the prevalent medical imaging practice to find abnormalities in soft tissues. Traditionally they are analyzed by a radiologist to detect abnormalities in soft tissues, especially the brain. The process of interpreting a massive volume of patient's MRI is laborious. Hence, the use of Machine Learning methodologies can aid in detecting abnormalities in soft tissues with considerable accuracy. In this research, we have curated a novel dataset and developed a framework that uses Deep Transfer Learning to perform a multi-classification of tumors in the brain MRI images. In this paper, we adopted the Deep Residual Convolutional Neural Network (ResNet50) architecture for the experiments along with discriminative learning techniques to train the model. Using the novel dataset and two publicly available MRI brain datasets, this proposed approach attained a classification accuracy of 86.40% on the curated dataset, 93.80% on the Harvard Whole Brain Atlas dataset, and 97.05% accuracy on the School of Biomedical Engineering dataset. Results of our experiments significantly demonstrate our proposed framework for transfer learning is a potential and effective method for brain tumor multi-classification tasks.
In this paper we present the results of our experiments in training and deploying a self-supervised retrieval-based chatbot trained with contrastive learning for assisting customer support agents. In contrast to most existing research papers in this area where the focus is on solving just one component of a deployable chatbot, we present an end-to-end set of solutions to take the reader from an unlabelled chatlogs to a deployed chatbot. This set of solutions includes creating a self-supervised dataset and a weakly labelled dataset from chatlogs, as well as a systematic approach to selecting a fixed list of canned responses. We present a hierarchical-based RNN architecture for the response selection model, chosen for its ability to cache intermediate utterance embeddings, which helped to meet deployment inference speed requirements. We compare the performance of this architecture across 3 different learning objectives: self-supervised contrastive learning, binary classification, and multi-class classification. We find that using a self-supervised contrastive learning model outperforms training the binary and multi-class classification models on a weakly labelled dataset. Our results validate that the self-supervised contrastive learning approach can be effectively used for a real-world chatbot scenario.
Many machine learning applications can benefit from simulated data for systematic validation - in particular if real-life data is difficult to obtain or annotate. However, since simulations are prone to domain shift w.r.t. real-life data, it is crucial to verify the transferability of the obtained results. We propose a novel framework consisting of a generative label-to-image synthesis model together with different transferability measures to inspect to what extent we can transfer testing results of semantic segmentation models from synthetic data to equivalent real-life data. With slight modifications, our approach is extendable to, e.g., general multi-class classification tasks. Grounded on the transferability analysis, our approach additionally allows for extensive testing by incorporating controlled simulations. We validate our approach empirically on a semantic segmentation task on driving scenes. Transferability is tested using correlation analysis of IoU and a learned discriminator. Although the latter can distinguish between real-life and synthetic tests, in the former we observe surprisingly strong correlations of 0.7 for both cars and pedestrians.
In forming learning objectives, one oftentimes needs to aggregate a set of individual values to a single output. Such cases occur in the aggregate loss, which combines individual losses of a learning model over each training sample, and in the individual loss for multi-label learning, which combines prediction scores over all class labels. In this work, we introduce the sum of ranked range (SoRR) as a general approach to form learning objectives. A ranked range is a consecutive sequence of sorted values of a set of real numbers. The minimization of SoRR is solved with the difference of convex algorithm (DCA). We explore two applications in machine learning of the minimization of the SoRR framework, namely the AoRR aggregate loss for binary/multi-class classification at the sample level and the TKML individual loss for multi-label/multi-class classification at the label level. A combination loss of AoRR and TKML is proposed as a new learning objective for improving the robustness of multi-label learning in the face of outliers in sample and labels alike. Our empirical results highlight the effectiveness of the proposed optimization frameworks and demonstrate the applicability of proposed losses using synthetic and real data sets.
Generalised linear models for multi-class classification problems are one of the fundamental building blocks of modern machine learning tasks. In this manuscript, we characterise the learning of a mixture of $K$ Gaussians with generic means and covariances via empirical risk minimisation (ERM) with any convex loss and regularisation. In particular, we prove exact asymptotics characterising the ERM estimator in high-dimensions, extending several previous results about Gaussian mixture classification in the literature. We exemplify our result in two tasks of interest in statistical learning: a) classification for a mixture with sparse means, where we study the efficiency of $\ell_1$ penalty with respect to $\ell_2$; b) max-margin multi-class classification, where we characterise the phase transition on the existence of the multi-class logistic maximum likelihood estimator for $K>2$. Finally, we discuss how our theory can be applied beyond the scope of synthetic data, showing that in different cases Gaussian mixtures capture closely the learning curve of classification tasks in real data sets.