This study investigates classifier performance across EEG frequency bands using various optimizers and evaluates efficient class prediction for the left and right hemispheres. Three neural network architectures - a deep dense network, a shallow three-layer network, and a convolutional neural network (CNN) - are implemented and compared using the TensorFlow and PyTorch frameworks. Results indicate that the Adagrad and RMSprop optimizers consistently perform well across different frequency bands, with Adadelta exhibiting robust performance in cross-model evaluations. Specifically, Adagrad excels in the beta band, while RMSprop achieves superior performance in the gamma band. Conversely, SGD and FTRL exhibit inconsistent performance. Among the models, the CNN demonstrates the second highest accuracy, particularly in capturing spatial features of EEG data. The deep dense network shows competitive performance in learning complex patterns, whereas the shallow three-layer network, sometimes being less accurate, provides computational efficiency. SHAP (Shapley Additive Explanations) plots are employed to identify efficient class prediction, revealing nuanced contributions of EEG frequency bands to model accuracy. Overall, the study highlights the importance of optimizer selection, model architecture, and EEG frequency band analysis in enhancing classifier performance and understanding feature importance in neuroimaging-based classification tasks.

Deep learning has emerged as the most promising approach in various fields; however, when the distributions of training and test data are different (domain shift), the performance of deep learning models can degrade. Semi-supervised domain adaptation (SSDA) is a major approach for addressing this issue, assuming that a fully labeled training set (source domain) is available, but the test set (target domain) provides labels only for a small subset. In this study, we propose a novel two-step momentum encoder-utilized SSDA framework, MoSSDA, for multivariate time-series classification. Time series data are highly sensitive to noise, and sequential dependencies cause domain shifts resulting in critical performance degradation. To obtain a robust, domain-invariant and class-discriminative representation, MoSSDA employs a domain-invariant encoder to learn features from both source and target domains. Subsequently, the learned features are fed to a mixup-enhanced positive contrastive module consisting of an online momentum encoder. The final classifier is trained with learned features that exhibit consistency and discriminability with limited labeled target domain data, without data augmentation. We applied a two-stage process by separating the gradient flow between the encoders and the classifier to obtain rich and complex representations. Through extensive experiments on six diverse datasets, MoSSDA achieved state-of-the-art performance for three different backbones and various unlabeled ratios in the target domain data. The Ablation study confirms that each module, including two-stage learning, is effective in improving the performance. Our code is available at https://github.com/seonyoungKimm/MoSSDA

Cloud computing offers on-demand resource access, regulated by Service-Level Agreements (SLAs) between consumers and Cloud Service Providers (CSPs). SLA violations can impact efficiency and CSP profitability. In this work, we propose an SLA-aware automated algorithm-selection framework for combinatorial optimization problems in resource-constrained cloud environments. The framework uses an ensemble of machine learning models to predict performance and rank algorithm-hardware pairs based on SLA constraints. We also apply our framework to the 0-1 knapsack problem. We curate a dataset comprising instance specific features along with memory usage, runtime, and optimality gap for 6 algorithms. As an empirical benchmark, we evaluate the framework on both classification and regression tasks. Our ablation study explores the impact of hyperparameters, learning approaches, and large language models effectiveness in regression, and SHAP-based interpretability.

While ViTs have achieved across machine learning tasks, deploying them in real-world scenarios faces a critical challenge: generalizing under OoD shifts. A crucial research gap exists in understanding how to design ViT architectures, both manually and automatically, for better OoD generalization. To this end, we introduce OoD-ViT-NAS, the first systematic benchmark for ViTs NAS focused on OoD generalization. This benchmark includes 3000 ViT architectures of varying computational budgets evaluated on 8 common OoD datasets. Using this benchmark, we analyze factors contributing to OoD generalization. Our findings reveal key insights. First, ViT architecture designs significantly affect OoD generalization. Second, ID accuracy is often a poor indicator of OoD accuracy, highlighting the risk of optimizing ViT architectures solely for ID performance. Third, we perform the first study of NAS for ViTs OoD robustness, analyzing 9 Training-free NAS methods. We find that existing Training-free NAS methods are largely ineffective in predicting OoD accuracy despite excelling at ID accuracy. Simple proxies like Param or Flop surprisingly outperform complex Training-free NAS methods in predicting OoD accuracy. Finally, we study how ViT architectural attributes impact OoD generalization and discover that increasing embedding dimensions generally enhances performance. Our benchmark shows that ViT architectures exhibit a wide range of OoD accuracy, with up to 11.85% improvement for some OoD shifts. This underscores the importance of studying ViT architecture design for OoD. We believe OoD-ViT-NAS can catalyze further research into how ViT designs influence OoD generalization.

Large Language Models (LLMs) have recently emerged, attracting considerable attention due to their ability to generate highly natural, human-like text. This study compares the latent community structures of LLM-generated text and human-written text within a hypothesis testing procedure. Specifically, we analyze three text sets: original human-written texts ($\mathcal{O}$), their LLM-paraphrased versions ($\mathcal{G}$), and a twice-paraphrased set ($\mathcal{S}$) derived from $\mathcal{G}$. Our analysis addresses two key questions: (1) Is the difference in latent community structures between $\mathcal{O}$ and $\mathcal{G}$ the same as that between $\mathcal{G}$ and $\mathcal{S}$? (2) Does $\mathcal{G}$ become more similar to $\mathcal{O}$ as the LLM parameter controlling text variability is adjusted? The first question is based on the assumption that if LLM-generated text truly resembles human language, then the gap between the pair ($\mathcal{O}$, $\mathcal{G}$) should be similar to that between the pair ($\mathcal{G}$, $\mathcal{S}$), as both pairs consist of an original text and its paraphrase. The second question examines whether the degree of similarity between LLM-generated and human text varies with changes in the breadth of text generation. To address these questions, we propose a statistical hypothesis testing framework that leverages the fact that each text has corresponding parts across all datasets due to their paraphrasing relationship. This relationship enables the mapping of one dataset's relative position to another, allowing two datasets to be mapped to a third dataset. As a result, both mapped datasets can be quantified with respect to the space characterized by the third dataset, facilitating a direct comparison between them. Our results indicate that GPT-generated text remains distinct from human-authored text.

Drug-target affinity (DTA) prediction is a critical aspect of drug discovery. The meaningful representation of drugs and targets is crucial for accurate prediction. Using 1D string-based representations for drugs and targets is a common approach that has demonstrated good results in drug-target affinity prediction. However, these approach lacks information on the relative position of the atoms and bonds. To address this limitation, graph-based representations have been used to some extent. However, solely considering the structural aspect of drugs and targets may be insufficient for accurate DTA prediction. Integrating the functional aspect of these drugs at the genetic level can enhance the prediction capability of the models. To fill this gap, we propose GramSeq-DTA, which integrates chemical perturbation information with the structural information of drugs and targets. We applied a Grammar Variational Autoencoder (GVAE) for drug feature extraction and utilized two different approaches for protein feature extraction: Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN). The chemical perturbation data is obtained from the L1000 project, which provides information on the upregulation and downregulation of genes caused by selected drugs. This chemical perturbation information is processed, and a compact dataset is prepared, serving as the functional feature set of the drugs. By integrating the drug, gene, and target features in the model, our approach outperforms the current state-of-the-art DTA prediction models when validated on widely used DTA datasets (BindingDB, Davis, and KIBA). This work provides a novel and practical approach to DTA prediction by merging the structural and functional aspects of biological entities, and it encourages further research in multi-modal DTA prediction.

In this paper, we propose a deep learning based system for the task of deepfake audio detection. In particular, the draw input audio is first transformed into various spectrograms using three transformation methods of Short-time Fourier Transform (STFT), Constant-Q Transform (CQT), Wavelet Transform (WT) combined with different auditory-based filters of Mel, Gammatone, linear filters (LF), and discrete cosine transform (DCT). Given the spectrograms, we evaluate a wide range of classification models based on three deep learning approaches. The first approach is to train directly the spectrograms using our proposed baseline models of CNN-based model (CNN-baseline), RNN-based model (RNN-baseline), C-RNN model (C-RNN baseline). Meanwhile, the second approach is transfer learning from computer vision models such as ResNet-18, MobileNet-V3, EfficientNet-B0, DenseNet-121, SuffleNet-V2, Swint, Convnext-Tiny, GoogLeNet, MNASsnet, RegNet. In the third approach, we leverage the state-of-the-art audio pre-trained models of Whisper, Seamless, Speechbrain, and Pyannote to extract audio embeddings from the input spectrograms. Then, the audio embeddings are explored by a Multilayer perceptron (MLP) model to detect the fake or real audio samples. Finally, high-performance deep learning models from these approaches are fused to achieve the best performance. We evaluated our proposed models on ASVspoof 2019 benchmark dataset. Our best ensemble model achieved an Equal Error Rate (EER) of 0.03, which is highly competitive to top-performing systems in the ASVspoofing 2019 challenge. Experimental results also highlight the potential of selective spectrograms and deep learning approaches to enhance the task of audio deepfake detection.

Social media has experienced incredible growth over the last decade, both in its scale and importance as a form of communication. The nature of social media means that anyone can post anything they desire, putting forward any position, whether it is enlightening, repugnant or anywhere between. Depending on the forum, such posts can be visible to many millions of people. Different forums have different definitions of inappropriate content and different processes for identifying it, but the scale of the medium means that automated methods are an important part of this task. Hate-speech is an important aspect of this inappropriate content. Hate-speech is a subjective and complex term with no single definition, however. Irrespective of the definition of the term or the problem, it is clear that automated methods for detecting hate-speech are necessary in some circumstances. In such cases it is critical that the methods employed are accurate, effective, and efficient. A variety of methods have been explored for the hate speech detection task, including traditional classifiers [6, 17, 32, 41,57], deep learning-based classifiers [1,7,8,46,59], or the combination of both approaches [7,25,37].