In the current digital era, facial recognition systems offer significant utility and have been widely integrated into modern technological infrastructures; however, their widespread use has also raised serious privacy concerns, prompting regulations that mandate data removal upon request. Machine unlearning has emerged as a powerful solution to address this issue by selectively removing the influence of specific user data from trained models while preserving overall model performance. However, existing machine unlearning techniques largely depend on supervised techniques requiring identity labels, which are often unavailable in privacy-constrained situations or in large-scale, noisy datasets. To address this critical gap, we introduce CURE (Centroid-guided Unsupervised Representation Erasure), the first unsupervised unlearning framework for facial recognition systems that operates without the use of identity labels, effectively removing targeted samples while preserving overall performance. We also propose a novel metric, the Unlearning Efficiency Score (UES), which balances forgetting and retention stability, addressing shortcomings in the current evaluation metrics. CURE significantly outperforms unsupervised variants of existing unlearning methods. Additionally, we conducted quality-aware unlearning by designating low-quality images as the forget set, demonstrating its usability and benefits, and highlighting the role of image quality in machine unlearning. The full code can be found here: https://github.com/Shivam101s/CURE_FaceUnlearning

Material property prediction is a critical task within material science and other related fields where the identification of advanced materials with dynamic characteristics is essential for developing innovative technologies, improving product performance, and driving scientific discoveries. The search for suitable candidate materials aimed at specific applications is traditionally conducted through time-consuming experimental methods. Conversely, computational approaches, such as density functional theory (DFT) calculations, necessitate the resolution of complex mathematical equations and significant computational resources. Machine learning (ML) models have significantly transformed this field by automating the tedious process of material searching within extensive search spaces. However, these models are unable to incorporate diverse features of varying dimensions for different materials and properties. Recently, graph convolutional neural networks (GCNNs), have been employed to analyze the complex structures of materials, thereby efficiently predicting their associated properties. In this work, we demonstrate the potential of multi-task learning (MTL) in conjunction with the GCNNs in material property prediction. Our findings suggest that the MTL framework integrated into the GCNN architecture such as Crystal Graph Convolutional Neural Network (CGCNN) and its advanced variant Orbital Graph Convolutional Neural Network (OGCNN) can enhance the generalization and efficacy of property predictions. Notably, we achieved a maximum improvement of 7.95 % in one of our setups, underscoring the potential of these models to handle multiple property predictions with considerable effectiveness.

The study of Maya hieroglyphic writing unlocks the rich history of cultural and societal knowledge embedded within this ancient civilization's visual narrative. Machine learning (ML) offers a novel lens through which we can translate these inscriptions, with the potential to allow non-specialists access to reading these texts and to aid in the decipherment of those hieroglyphs which continue to elude comprehensive interpretation. Toward this, we leverage a large foundation model to segment Maya hieroglyphs from an open-source digital library dedicated to Maya artifacts. Despite the initial promise of publicly available foundation segmentation models, their effectiveness in accurately segmenting Maya hieroglyphs was initially limited. Addressing this challenge, our study involved the meticulous curation of image and label pairs with the assistance of experts in Maya art and history, enabling the fine-tuning of these foundation models. This process significantly enhanced model performance, illustrating the potential of fine-tuning approaches and the value of our expanding dataset. We plan to open-source this dataset for encouraging future research, and eventually to help make the hieroglyphic texts legible to a broader community, particularly for Maya heritage community members.

The success of deep learning relies on the availability of large-scale annotated data sets, the acquisition of which can be costly requiring expert domain knowledge. Semi-supervised learning (SSL) mitigates this challenge by exploiting the behavior of the neural function on large unlabeled data. The smoothness of the neural function is a commonly used assumption exploited in SSL. A successful example is the adoption of mixup strategy in SSL that enforces the global smoothness of the neural function by encouraging it to behave linearly when interpolating between training examples. Despite its empirical success, however, the theoretical underpinning of how mixup regularizes the neural function has not been fully understood. In this paper, we offer a theoretically substantiated proposition that mixup improves the smoothness of the neural function by bounding the Lipschitz constant of the gradient function of the neural networks. We then propose that this can be strengthened by simultaneously constraining the Lipschitz constant of the neural function itself through adversarial Lipschitz regularization, encouraging the neural function to behave linearly while also constraining the slope of this linear function. On three benchmark data sets and one real-world biomedical data set, we demonstrate that this combined regularization results in improved generalization performance of SSL when learning from a small amount of labeled data. Our code is available at https://github.com/Prasanna1991/Mixup-LR.

To improve the ability of variational auto-encoders (VAE) to disentangle in the latent space, existing works mostly focus on enforcing the independence among the learned latent factors. However, the ability of these models to disentangle often decreases as the complexity of the generative factors increases. In this paper, we investigate the little-explored effect of the modeling capacity of a posterior density on the disentangling ability of the VAE. We note that the independence within and the complexity of the latent density are two different properties we constrain when regularizing the posterior density: while the former promotes the disentangling ability of VAE, the latter - if overly limited - creates an unnecessary competition with the data reconstruction objective in VAE. Therefore, if we preserve the independence but allow richer modeling capacity in the posterior density, we will lift this competition and thereby allow improved independence and data reconstruction at the same time. We investigate this theoretical intuition with a VAE that utilizes a non-parametric latent factor model, the Indian Buffet Process (IBP), as a latent density that is able to grow with the complexity of the data. Across two widely-used benchmark data sets (MNIST and dSprites) and two clinical data sets little explored for disentangled learning, we qualitatively and quantitatively demonstrated the improved disentangling performance of IBP-VAE over the state of the art. In the latter two clinical data sets riddled with complex factors of variations, we further demonstrated that unsupervised disentangling of nuisance factors via IBP-VAE - when combined with a supervised objective - can not only improve task accuracy in comparison to relevant supervised deep architectures, but also facilitate knowledge discovery related to task decision-making.