Deep Neural Networks (DNNs) trained on one dataset (source domain) do not perform well on another set of data (target domain), which is different but has similar properties as the source domain. Domain Adaptation (DA) strives to alleviate this problem and has great potential in its application in practical settings, real-world scenarios, industrial applications and many data domains. Various DA methods aimed at individual data domains have been reported in the last few years; however, there is no comprehensive survey that encompasses all these data domains, focuses on the datasets available, the methods relevant to each domain, and importantly the applications and challenges. To that end, this survey paper discusses how DA can help DNNs work efficiently in these settings by reviewing DA methods and techniques. We have considered five data domains: computer vision, natural language processing, speech, time-series, and multi-modal data. We present a comprehensive taxonomy, including the methods, datasets, challenges, and applications corresponding to each domain. Our goal is to discuss industrial use cases and DA implementation for those. Our final aim is to provide future research directions based on evolving methods and results, the datasets used, and industrial applications.

Automatic emotion detection has become crucial in various domains, such as healthcare, neuroscience, smart home technologies, and human-computer interaction (HCI). Speech Emotion Recognition (SER) has attracted considerable attention because of its potential to improve conversational robotics and human-computer interaction (HCI) systems. Despite its promise, SER research faces challenges such as data scarcity, the subjective nature of emotions, and complex feature extraction methods. In this paper, we seek to investigate whether a lightweight deep neural ensemble model (CNN and CNN_Bi-LSTM) using well-known hand-crafted features such as ZCR, RMSE, Chroma STFT, and MFCC would outperform models that use automatic feature extraction techniques (e.g., spectrogram-based methods) on benchmarked datasets. The focus of this paper is on the effectiveness of careful fine-tuning of the neural models with learning rate (LR) schedulers and applying regularization techniques. Our proposed ensemble model is validated using five publicly available datasets: RAVDESS, TESS, SAVEE, CREMA-D, and EmoDB. Accuracy, AUC-ROC, AUC-PRC, and F1-score metrics were used for performance testing, and the LIME (Local Interpretable Model-agnostic Explanations) technique was used for interpreting the results of our proposed ensemble model. Results indicate that our ensemble model consistently outperforms individual models, as well as several compared models which include spectrogram-based models for the above datasets in terms of the evaluation metrics.

Persistent homology is a powerful tool in topological data analysis (TDA) to compute, study, and encode efficiently multi-scale topological features and is being increasingly used in digital image classification. The topological features represent a number of connected components, cycles, and voids that describe the shape of data. Persistent homology extracts the birth and death of these topological features through a filtration process. The lifespan of these features can be represented using persistent diagrams (topological signatures). Cubical homology is a more efficient method for extracting topological features from a 2D image and uses a collection of cubes to compute the homology, which fits the digital image structure of grids. In this research, we propose a cubical homology-based algorithm for extracting topological features from 2D images to generate their topological signatures. Additionally, we propose a novel score measure, which measures the significance of each of the sub-simplices in terms of persistence. In addition, gray-level co-occurrence matrix (GLCM) and contrast limited adapting histogram equalization (CLAHE) are used as supplementary methods for extracting features. Supervised machine learning models are trained on selected image datasets to study the efficacy of the extracted topological features. Among the eight tested models with six published image datasets of varying pixel sizes, classes, and distributions, our experiments demonstrate that cubical homology-based machine learning with the deep residual network (ResNet 1D) and Light Gradient Boosting Machine (lightGBM) shows promise with the extracted topological features.