Swinburne University of Technology

About: Swinburne University of Technology is a education organization based out in Melbourne, Victoria, Australia. It is known for research contribution in the topics: Galaxy & Population. The organization has 7223 authors who have published 25530 publications receiving 667955 citations. The organization is also known as: Swinburne Technical College & Swinburne College of Technology.

Vitamins and Minerals for Energy, Fatigue and Cognition: A Narrative Review of the Biochemical and Clinical Evidence

Abstract: Vitamins and minerals are essential to humans as they play essential roles in a variety of basic metabolic pathways that support fundamental cellular functions. In particular, their involvement in energy-yielding metabolism, DNA synthesis, oxygen transport, and neuronal functions makes them critical for brain and muscular function. These, in turn, translate into effects on cognitive and psychological processes, including mental and physical fatigue. This review is focused on B vitamins (B1, B2, B3, B5, B6, B8, B9 and B12), vitamin C, iron, magnesium and zinc, which have recognized roles in these outcomes. It summarizes the biochemical bases and actions of these micronutrients at both the molecular and cellular levels and connects them with cognitive and psychological symptoms, as well as manifestations of fatigue that may occur when status or supplies of these micronutrients are not adequate.

Transformation from spirals into S0s with bulge growth in groups of galaxies

Abstract: Recent observations have revealed that the time evolution of the S0 number fraction at intermediate and high redshifts (0.2 < z < 0.8) is more dramatic in groups of galaxies than in clusters. In order to understand the origin of S0s in groups, we investigate numerically the morphological transformation of spirals into S0s through group-related physical processes. Our chemodynamical simulations show that spirals in group environments can be strongly influenced by repetitive slow encounters with group member galaxies so that those with thin disks and prominent spiral arm structures can be transformed into S0s with thick disks and without prominent spiral arm structure. Such tidal interactions can also trigger repetitive starbursts within the bulges of spirals and consequently increase significantly the masses of their bulges. Owing to rapid consumption of gas initially in spirals during the bulge growth, the S0s can become gas-poor. The S0s transformed from spirals in this way have young and metal-rich stellar populations in the inner regions of their bulges. The simulated S0s have lower maximum rotational velocities and flatter radial line-of-sight velocity dispersion profiles in comparison to their progenitor spirals. The formation processes of S0s due to tidal interactions depend not only on the masses and orbits of the progenitor spirals, but also on group mass. A significant fraction (10 30%) of stars and gas can be stripped during this spiral to S0 morphological transformation so that intragroup stars and gas can be formed. Based on these results, we discuss structures, kinematics, chemical properties, and the Tully-Fisher relation of S0s in groups.

Towards Fair and Privacy-Preserving Federated Deep Models

Abstract: The current standalone deep learning framework tends to result in overfitting and low utility. This problem can be addressed by either a centralized framework that deploys a central server to train a global model on the joint data from all parties, or a distributed framework that leverages a parameter server to aggregate local model updates. Server-based solutions are prone to the problem of a single-point-of-failure. In this respect, collaborative learning frameworks, such as federated learning (FL), are more robust. Existing federated learning frameworks overlook an important aspect of participation: fairness. All parties are given the same final model without regard to their contributions. To address these issues, we propose a decentralized Fair and Privacy-Preserving Deep Learning (FPPDL) framework to incorporate fairness into federated deep learning models. In particular, we design a local credibility mutual evaluation mechanism to guarantee fairness, and a three-layer onion-style encryption scheme to guarantee both accuracy and privacy. Different from existing FL paradigm, under FPPDL, each participant receives a different version of the FL model with performance commensurate with his contributions. Experiments on benchmark datasets demonstrate that FPPDL balances fairness, privacy and accuracy. It enables federated learning ecosystems to detect and isolate low-contribution parties, thereby promoting responsible participation.