The Evidential Regression Network (ERN) represents a novel approach that integrates deep learning with Dempster-Shafer's theory to predict a target and quantify the associated uncertainty. Guided by the underlying theory, specific activation functions must be employed to enforce non-negative values, which is a constraint that compromises model performance by limiting its ability to learn from all samples. This paper provides a theoretical analysis of this limitation and introduces an improvement to overcome it. Initially, we define the region where the models can't effectively learn from the samples. Following this, we thoroughly analyze the ERN and investigate this constraint. Leveraging the insights from our analysis, we address the limitation by introducing a novel regularization term that empowers the ERN to learn from the whole training set. Our extensive experiments substantiate our theoretical findings and demonstrate the effectiveness of the proposed solution.

Scientists have discovered a hidden layer of Earth, which sits 100 miles below the surface and covers at least 44 percent of the planet. This previously unknown region of molten rock is part of the asthenosphere, located under tectonic plates in the upper mantle, which forms a soft boundary that allows the solid rock slabs to move. While the discovery is significant, it shatters long-held theories that molten rocks influence the asthenosphere's viscosity. Junlin Hua, with the University of Texas, Austin, said in a statement: 'When we think about something melting, we intuitively think that the melt must play a big role in the material's viscosity. 'But what we found is that even where the melt fraction is quite high, its effect on mantle flow is very minor.'

Matrix factorization (MF) is a prevailing collaborative filtering method for building recommender systems. It requires users to upload their personal preferences to the recommender for performing MF, which raises serious privacy concerns. This paper proposes a differentially private MF mechanism that can prevent an untrusted recommender from learning any users' ratings or profiles. Our design decouples computations upon users' private data from the recommender to users, and makes the recommender aggregate local results in a privacy-preserving way. It uses the objective perturbation to make sure that the final item profiles satisfy differential privacy and solves the challenge to decompose the noise component for objective perturbation into small pieces that can be determined locally and independently by users. We also propose a third-party based mechanism to reduce noises added in each iteration and adapt our online algorithm to the dynamic setting that allows users to leave and join. The experiments show that our proposal is efficient and introduces acceptable side effects on the precision of results.

Michelle Hua took home the top spot in a major international science competition. Hua is a sophomore at Cranbrook Kingswood School. She found out on Friday that she was a top winner of the Regeneron International Science and Engineering Fair. Judges deemed Hua's project a fundamental contribution to the field of machine learning with applications in artificial intelligence. She is the first female to win the top prize in a decade.