The field of machine learning is moving towards developing more robust and reliable methods for learning from noisy and uncertain data. Recent research has focused on improving the resilience of deep neural networks to label noise, data corruption, and other forms of uncertainty. One key direction is the development of novel loss functions and regularization techniques that can adaptively handle noisy labels and uncertain data. Another important area of research is active learning, where the goal is to select the most informative samples for labeling in order to achieve better performance with limited labeled data. Notable papers in this area include those that propose new frameworks for active learning, such as oracle-free active learning schemes, and those that develop more robust loss functions for learning with noisy labels. Some papers that are particularly noteworthy include: Selection-Based Vulnerabilities: Clean-Label Backdoor Attacks in Active Learning, which introduces a practical framework to reveal the weakness of active learning. Introducing Fractional Classification Loss for Robust Learning with Noisy Labels, which proposes an adaptive robust loss that automatically calibrates its robustness to label noise during training. Learning to Forget with Information Divergence Reweighted Objectives for Noisy Labels, which introduces a new class of objectives for learning under noisy labels. OFAL: An Oracle-Free Active Learning Framework, which utilizes neural network uncertainty to transform highly confident unlabeled samples into informative uncertain samples. Adaptive Confidence-Wise Loss for Improved Lens Structure Segmentation in AS-OCT, which proposes an adaptive confidence-wise loss to group each lens structure sub-region into different confidence sub-regions. Image selective encryption analysis using mutual information in CNN based embedding space, which examines information leakage in image data. RL-MoE: An Image-Based Privacy Preserving Approach In Intelligent Transportation System, which proposes a novel framework that transforms sensitive visual data into privacy-preserving textual descriptions. Combating Noisy Labels via Dynamic Connection Masking, which proposes a dynamic connection masking mechanism for both Multi-Layer Perceptron Networks and KANs to enhance the robustness of classifiers against noisy labels.