As the world continues to face the challenges of climate change, it is crucial to consider the environmental impact of the technologies we use. In this study, we investigate the performance and computational carbon emissions of various transfer learning models for garbage classification. We examine the MobileNet, ResNet50, ResNet101, and EfficientNetV2S and EfficientNetV2M models. Our findings indicate that the EfficientNetV2 family achieves the highest accuracy, recall, f1-score, and IoU values. However, the EfficientNetV2M model requires more time and produces higher carbon emissions. ResNet50 outperforms ResNet110 in terms of accuracy, recall, f1-score, and IoU, but it has a larger carbon footprint. We conclude that EfficientNetV2S is the most sustainable and accurate model with 96.41% accuracy. Our research highlights the significance of considering the ecological impact of machine learning models in garbage classification.
This study assesses deep learning models for audio classification in a clinical setting with the constraint of small datasets reflecting real-world prospective data collection. We analyze CNNs, including DenseNet and ConvNeXt, alongside transformer models like ViT, SWIN, and AST, and compare them against pre-trained audio models such as YAMNet and VGGish. Our method highlights the benefits of pre-training on large datasets before fine-tuning on specific clinical data. We prospectively collected two first-of-their-kind patient audio datasets from stroke patients. We investigated various preprocessing techniques, finding that RGB and grayscale spectrogram transformations affect model performance differently based on the priors they learn from pre-training. Our findings indicate CNNs can match or exceed transformer models in small dataset contexts, with DenseNet-Contrastive and AST models showing notable performance. This study highlights the significance of incremental marginal gains through model selection, pre-training, and preprocessing in sound classification; this offers valuable insights for clinical diagnostics that rely on audio classification.
Graph Neural Networks (GNNs) have been widely used to learn node representations and with outstanding performance on various tasks such as node classification. However, noise, which inevitably exists in real-world graph data, would considerably degrade the performance of GNNs revealed by recent studies. In this work, we propose a novel and robust GNN encoder, Low-Rank Graph Contrastive Learning (LR-GCL). Our method performs transductive node classification in two steps. First, a low-rank GCL encoder named LR-GCL is trained by prototypical contrastive learning with low-rank regularization. Next, using the features produced by LR-GCL, a linear transductive classification algorithm is used to classify the unlabeled nodes in the graph. Our LR-GCL is inspired by the low frequency property of the graph data and its labels, and it is also theoretically motivated by our sharp generalization bound for transductive learning. To the best of our knowledge, our theoretical result is among the first to theoretically demonstrate the advantage of low-rank learning in graph contrastive learning supported by strong empirical performance. Extensive experiments on public benchmarks demonstrate the superior performance of LR-GCL and the robustness of the learned node representations. The code of LR-GCL is available at \url{https://anonymous.4open.science/r/Low-Rank_Graph_Contrastive_Learning-64A6/}.
We consider the problem of Neyman-Pearson classification which models unbalanced classification settings where error w.r.t. a distribution $\mu_1$ is to be minimized subject to low error w.r.t. a different distribution $\mu_0$. Given a fixed VC class $\mathcal{H}$ of classifiers to be minimized over, we provide a full characterization of possible distribution-free rates, i.e., minimax rates over the space of all pairs $(\mu_0, \mu_1)$. The rates involve a dichotomy between hard and easy classes $\mathcal{H}$ as characterized by a simple geometric condition, a three-points-separation condition, loosely related to VC dimension.
Quantum machine learning provides a fundamentally different approach to analyzing data. However, many interesting datasets are too complex for currently available quantum computers. Present quantum machine learning applications usually diminish this complexity by reducing the dimensionality of the data, e.g., via auto-encoders, before passing it through the quantum models. Here, we design a classical-quantum paradigm that unifies the dimensionality reduction task with a quantum classification model into a single architecture: the guided quantum compression model. We exemplify how this architecture outperforms conventional quantum machine learning approaches on a challenging binary classification problem: identifying the Higgs boson in proton-proton collisions at the LHC. Furthermore, the guided quantum compression model shows better performance compared to the deep learning benchmark when using solely the kinematic variables in our dataset.
A prominent family of methods for learning data distributions relies on density ratio estimation (DRE), where a model is trained to $\textit{classify}$ between data samples and samples from some reference distribution. DRE-based models can directly output the likelihood for any given input, a highly desired property that is lacking in most generative techniques. Nevertheless, to date, DRE methods have failed in accurately capturing the distributions of complex high-dimensional data, like images, and have thus been drawing reduced research attention in recent years. In this work we present $\textit{classification diffusion models}$ (CDMs), a DRE-based generative method that adopts the formalism of denoising diffusion models (DDMs) while making use of a classifier that predicts the level of noise added to a clean signal. Our method is based on an analytical connection that we derive between the MSE-optimal denoiser for removing white Gaussian noise and the cross-entropy-optimal classifier for predicting the noise level. Our method is the first DRE-based technique that can successfully generate images beyond the MNIST dataset. Furthermore, it can output the likelihood of any input in a single forward pass, achieving state-of-the-art negative log likelihood (NLL) among methods with this property. Code is available on the project's webpage in https://shaharYadin.github.io/CDM/ .
This paper presents a hand shape classification approach employing multiscale template matching. The integration of background subtraction is utilized to derive a binary image of the hand object, enabling the extraction of key features such as centroid and bounding box. The methodology, while simple, demonstrates effectiveness in basic hand shape classification tasks, laying the foundation for potential applications in straightforward human-computer interaction scenarios. Experimental results highlight the system's capability in controlled environments.
In strategic classification, agents modify their features, at a cost, to ideally obtain a positive classification from the learner's classifier. The typical response of the learner is to carefully modify their classifier to be robust to such strategic behavior. When reasoning about agent manipulations, most papers that study strategic classification rely on the following strong assumption: agents fully know the exact parameters of the deployed classifier by the learner. This often is an unrealistic assumption when using complex or proprietary machine learning techniques in real-world prediction tasks. We initiate the study of partial information release by the learner in strategic classification. We move away from the traditional assumption that agents have full knowledge of the classifier. Instead, we consider agents that have a common distributional prior on which classifier the learner is using. The learner in our model can reveal truthful, yet not necessarily complete, information about the deployed classifier to the agents. The learner's goal is to release just enough information about the classifier to maximize accuracy. We show how such partial information release can, counter-intuitively, benefit the learner's accuracy, despite increasing agents' abilities to manipulate. We show that while it is intractable to compute the best response of an agent in the general case, there exist oracle-efficient algorithms that can solve the best response of the agents when the learner's hypothesis class is the class of linear classifiers, or when the agents' cost function satisfies a natural notion of submodularity as we define. We then turn our attention to the learner's optimization problem and provide both positive and negative results on the algorithmic problem of how much information the learner should release about the classifier to maximize their expected accuracy.
Spinal metastasis is the most common disease in bone metastasis and may cause pain, instability and neurological injuries. Early detection of spinal metastasis is critical for accurate staging and optimal treatment. The diagnosis is usually facilitated with Computed Tomography (CT) scans, which requires considerable efforts from well-trained radiologists. In this paper, we explore a learning-based automatic bone quality classification method for spinal metastasis based on CT images. We simultaneously take the posterolateral spine involvement classification task into account, and employ multi-task learning (MTL) technique to improve the performance. MTL acts as a form of inductive bias which helps the model generalize better on each task by sharing representations between related tasks. Based on the prior knowledge that the mixed type can be viewed as both blastic and lytic, we model the task of bone quality classification as two binary classification sub-tasks, i.e., whether blastic and whether lytic, and leverage a multiple layer perceptron to combine their predictions. In order to make the model more robust and generalize better, self-paced learning is adopted to gradually involve from easy to more complex samples into the training process. The proposed learning-based method is evaluated on a proprietary spinal metastasis CT dataset. At slice level, our method significantly outperforms an 121-layer DenseNet classifier in sensitivities by $+12.54\%$, $+7.23\%$ and $+29.06\%$ for blastic, mixed and lytic lesions, respectively, meanwhile $+12.33\%$, $+23.21\%$ and $+34.25\%$ at vertebrae level.
Graph Neural Networks (GNNs) have demonstrated significant success in learning from graph-structured data across various domains. Despite their great successful, one critical challenge is often overlooked by existing works, i.e., the learning of message propagation that can generalize effectively to underrepresented graph regions. These minority regions often exhibit irregular homophily/heterophily patterns and diverse neighborhood class distributions, resulting in ambiguity. In this work, we investigate the ambiguity problem within GNNs, its impact on representation learning, and the development of richer supervision signals to fight against this problem. We conduct a fine-grained evaluation of GNN, analyzing the existence of ambiguity in different graph regions and its relation with node positions. To disambiguate node embeddings, we propose a novel method, {\method}, which exploits additional optimization guidance to enhance representation learning, particularly for nodes in ambiguous regions. {\method} identifies ambiguous nodes based on temporal inconsistency of predictions and introduces a disambiguation regularization by employing contrastive learning in a topology-aware manner. {\method} promotes discriminativity of node representations and can alleviating semantic mixing caused by message propagation, effectively addressing the ambiguity problem. Empirical results validate the efficiency of {\method} and highlight its potential to improve GNN performance in underrepresented graph regions.
Deep neural networks are extensively applied to real-world tasks, such as face recognition and medical image classification, where privacy and data protection are critical. Image data, if not protected, can be exploited to infer personal or contextual information. Existing privacy preservation methods, like encryption, generate perturbed images that are unrecognizable to even humans. Adversarial attack approaches prohibit automated inference even for authorized stakeholders, limiting practical incentives for commercial and widespread adaptation. This pioneering study tackles an unexplored practical privacy preservation use case by generating human-perceivable images that maintain accurate inference by an authorized model while evading other unauthorized black-box models of similar or dissimilar objectives, and addresses the previous research gaps. The datasets employed are ImageNet, for image classification, Celeba-HQ dataset, for identity classification, and AffectNet, for emotion classification. Our results show that the generated images can successfully maintain the accuracy of a protected model and degrade the average accuracy of the unauthorized black-box models to 11.97%, 6.63%, and 55.51% on ImageNet, Celeba-HQ, and AffectNet datasets, respectively.
Covariance and Hessian matrices have been analyzed separately in the literature for classification problems. However, integrating these matrices has the potential to enhance their combined power in improving classification performance. We present a novel approach that combines the eigenanalysis of a covariance matrix evaluated on a training set with a Hessian matrix evaluated on a deep learning model to achieve optimal class separability in binary classification tasks. Our approach is substantiated by formal proofs that establish its capability to maximize between-class mean distance (the concept of \textit{separation}) and minimize within-class variances (the concept of \textit{compactness}), which together define the two linear discriminant analysis (LDA) criteria, particularly under ideal data conditions such as isotropy around class means and dominant leading eigenvalues. By projecting data into the combined space of the most relevant eigendirections from both matrices, we achieve optimal class separability as per these LDA criteria. Empirical validation across neural and health datasets consistently supports our theoretical framework and demonstrates that our method outperforms established methods. Our method stands out by addressing both separation and compactness criteria, unlike PCA and the Hessian method, which predominantly emphasize one criterion each. This comprehensive approach captures intricate patterns and relationships, enhancing classification performance. Furthermore, through the utilization of both LDA criteria, our method outperforms LDA itself by leveraging higher-dimensional feature spaces, in accordance with Cover's theorem, which favors linear separability in higher dimensions. Additionally, our approach sheds light on complex DNN decision-making, rendering them comprehensible within a 2D space.
We propose a novel semiparametric classifier based on Mahalanobis distances of an observation from the competing classes. Our tool is a generalized additive model with the logistic link function that uses these distances as features to estimate the posterior probabilities of different classes. While popular parametric classifiers like linear and quadratic discriminant analyses are mainly motivated by the normality of the underlying distributions, the proposed classifier is more flexible and free from such parametric modeling assumptions. Since the densities of elliptic distributions are functions of Mahalanobis distances, this classifier works well when the competing classes are (nearly) elliptic. In such cases, it often outperforms popular nonparametric classifiers, especially when the sample size is small compared to the dimension of the data. To cope with non-elliptic and possibly multimodal distributions, we propose a local version of the Mahalanobis distance. Subsequently, we propose another classifier based on a generalized additive model that uses the local Mahalanobis distances as features. This nonparametric classifier usually performs like the Mahalanobis distance based semiparametric classifier when the underlying distributions are elliptic, but outperforms it for several non-elliptic and multimodal distributions. We also investigate the behaviour of these two classifiers in high dimension, low sample size situations. A thorough numerical study involving several simulated and real datasets demonstrate the usefulness of the proposed classifiers in comparison to many state-of-the-art methods.
This paper studies online structured prediction with full-information feedback. For online multiclass classification, Van der Hoeven (2020) established \emph{finite} surrogate regret bounds, which are independent of the time horizon, by introducing an elegant \emph{exploit-the-surrogate-gap} framework. However, this framework has been limited to multiclass classification primarily because it relies on a classification-specific procedure for converting estimated scores to outputs. We extend the exploit-the-surrogate-gap framework to online structured prediction with \emph{Fenchel--Young losses}, a large family of surrogate losses that includes the logistic loss for multiclass classification as a special case, obtaining finite surrogate regret bounds in various structured prediction problems. To this end, we propose and analyze \emph{randomized decoding}, which converts estimated scores to general structured outputs. Moreover, by applying our decoding to online multiclass classification with the logistic loss, we obtain a surrogate regret bound of $O(\| \mathbf{U} \|_\mathrm{F}^2)$, where $\mathbf{U}$ is the best offline linear estimator and $\| \cdot \|_\mathrm{F}$ denotes the Frobenius norm. This bound is tight up to logarithmic factors and improves the previous bound of $O(d\| \mathbf{U} \|_\mathrm{F}^2)$ due to Van der Hoeven (2020) by a factor of $d$, the number of classes.
Graph learning is crucial in the fields of bioinformatics, social networks, and chemicals. Although high-order graphlets, such as cycles, are critical to achieving an informative graph representation for node classification, edge prediction, and graph recognition, modeling high-order topological characteristics poses significant computational challenges, restricting its widespread applications in machine learning. To address this limitation, we introduce the concept of \textit{message detouring} to hierarchically characterize cycle representation throughout the entire graph, which capitalizes on the contrast between the shortest and longest pathways within a range of local topologies associated with each graph node. The topological feature representations derived from our message detouring landscape demonstrate comparable expressive power to high-order \textit{Weisfeiler-Lehman} (WL) tests but much less computational demands. In addition to the integration with graph kernel and message passing neural networks, we present a novel message detouring neural network, which uses Transformer backbone to integrate cycle representations across nodes and edges. Aside from theoretical results, experimental results on expressiveness, graph classification, and node classification show message detouring can significantly outperform current counterpart approaches on various benchmark datasets.
The value of text classification's future research has encountered challenges and uncertainties, due to the extraordinary efficacy demonstrated by large language models (LLMs) across numerous downstream NLP tasks. In this era of open-ended language modeling, where task boundaries are gradually fading, an urgent question emerges: have we made significant advances in text classification under the full benefit of LLMs? To answer this question, we propose RGPT, an adaptive boosting framework tailored to produce a specialized text classification LLM by recurrently ensembling a pool of strong base learners. The base learners are constructed by adaptively adjusting the distribution of training samples and iteratively fine-tuning LLMs with them. Such base learners are then ensembled to be a specialized text classification LLM, by recurrently incorporating the historical predictions from the previous learners. Through a comprehensive empirical comparison, we show that RGPT significantly outperforms 8 SOTA PLMs and 7 SOTA LLMs on four benchmarks by 1.36% on average. Further evaluation experiments show a clear surpassing of RGPT over human classification.
Consider the domain of multiclass classification within the adversarial online setting. What is the price of relying on bandit feedback as opposed to full information? To what extent can an adaptive adversary amplify the loss compared to an oblivious one? To what extent can a randomized learner reduce the loss compared to a deterministic one? We study these questions in the mistake bound model and provide nearly tight answers. We demonstrate that the optimal mistake bound under bandit feedback is at most $O(k)$ times higher than the optimal mistake bound in the full information case, where $k$ represents the number of labels. This bound is tight and provides an answer to an open question previously posed and studied by Daniely and Helbertal ['13] and by Long ['17, '20], who focused on deterministic learners. Moreover, we present nearly optimal bounds of $\tilde{\Theta}(k)$ on the gap between randomized and deterministic learners, as well as between adaptive and oblivious adversaries in the bandit feedback setting. This stands in contrast to the full information scenario, where adaptive and oblivious adversaries are equivalent, and the gap in mistake bounds between randomized and deterministic learners is a constant multiplicative factor of $2$. In addition, our results imply that in some cases the optimal randomized mistake bound is approximately the square-root of its deterministic parallel. Previous results show that this is essentially the smallest it can get.
Large occlusions result in a significant decline in image classification accuracy. During inference, diverse types of unseen occlusions introduce out-of-distribution data to the classification model, leading to accuracy dropping as low as 50%. As occlusions encompass spatially connected regions, conventional methods involving feature reconstruction are inadequate for enhancing classification performance. We introduce LEARN: Latent Enhancing feAture Reconstruction Network -- An auto-encoder based network that can be incorporated into the classification model before its classifier head without modifying the weights of classification model. In addition to reconstruction and classification losses, training of LEARN effectively combines intra- and inter-class losses calculated over its latent space -- which lead to improvement in recovering latent space of occluded data, while preserving its class-specific discriminative information. On the OccludedPASCAL3D+ dataset, the proposed LEARN outperforms standard classification models (VGG16 and ResNet-50) by a large margin and up to 2% over state-of-the-art methods. In cross-dataset testing, our method improves the average classification accuracy by more than 5% over the state-of-the-art methods. In every experiment, our model consistently maintains excellent accuracy on in-distribution data.
Data augmentation has been pivotal in successfully training deep learning models on classification tasks over the past decade. An important subclass of data augmentation techniques - which includes both label smoothing and Mixup - involves modifying not only the input data but also the input label during model training. In this work, we analyze the role played by the label augmentation aspect of such methods. We first prove that linear models on binary classification data trained with label augmentation learn only the minimum variance features in the data, while standard training (which includes weight decay) can learn higher variance features. We then use our techniques to show that even for nonlinear models and general data distributions, the label smoothing and Mixup losses are lower bounded by a function of the model output variance. An important consequence of our results is negative: label smoothing and Mixup can be less robust to spurious correlations in the data. We verify that our theory reflects practice via experiments on image classification benchmarks modified to have spurious correlations.
As network security receives widespread attention, encrypted traffic classification has become the current research focus. However, existing methods conduct traffic classification without sufficiently considering the common characteristics between data samples, leading to suboptimal performance. Moreover, they train the packet-level and flow-level classification tasks independently, which is redundant because the packet representations learned in the packet-level task can be exploited by the flow-level task. Therefore, in this paper, we propose an effective model named a Contrastive Learning Enhanced Temporal Fusion Encoder (CLE-TFE). In particular, we utilize supervised contrastive learning to enhance the packet-level and flow-level representations and perform graph data augmentation on the byte-level traffic graph so that the fine-grained semantic-invariant characteristics between bytes can be captured through contrastive learning. We also propose cross-level multi-task learning, which simultaneously accomplishes the packet-level and flow-level classification tasks in the same model with one training. Further experiments show that CLE-TFE achieves the best overall performance on the two tasks, while its computational overhead (i.e., floating point operations, FLOPs) is only about 1/14 of the pre-trained model (e.g., ET-BERT). We release the code at https://github.com/ViktorAxelsen/CLE-TFE