Sex Differences in Genetic Effects Using Two Gene Aggregation Techniques: Depression and Cognition

The state of science has increasingly valued interdisciplinary training and research. As an early-stage investigator, interdisciplinary research, such as that supported by the Research Centers Collaborative Network through the National Institute on Aging, provides valuable opportunities to connect researchers across disciplinary boundaries. This program encourages researchers use their diverse training experiences to tackle difficult questions in aging research and forge scholarly relationship that advance the state of science through creative problem-solving and a spectrum of interdisciplinary perspectives. This presentation describes a recently funded project to examine sex differences in aging outcomes using genetics, depression, and impaired cognition with polygenic and gene-region aggregation techniques. We will highlight the benefits and opportunities of the Research Centers Collaborative Network pilot grant to this highly interdisciplinary work, co-lead by early-stage investigators.

Region-Based Global Reasoning Networks

Global reasoning plays a significant role in many computer vision tasks which need to capture long-distance relationships. However, most current studies on global reasoning focus on exploring the relationship between pixels and ignore the critical role of the regions. In this paper, we propose an novel approach that explores the relationship between regions which have richer semantics than pixels. Specifically, we design a region aggregation method that can gather regional features automatically into a uniform shape, and adjust theirs positions adaptively for better alignment. To achieve the best performance of global reasoning, we propose various relationship exploration methods and apply them on the regional features. Our region-based global reasoning module, named ReGr, is end-to-end and can be inserted into existing visual understanding models without extra supervision. To evaluate our approach, we apply ReGr to fine-grained classification and action recognition benchmark tasks, and the experimental results demonstrate the effectiveness of our approach.

INDUCED AND LOGARITHMIC DISTANCES WITH MULTI-REGION AGGREGATION OPERATORS

This paper introduces the induced ordered weighted logarithmic averaging IOWLAD and multiregion induced ordered weighted logarithmic averaging MR-IOWLAD operators. The distinctive characteristic of these operators lies in the notion of distance measures combined with the complex reordering mechanism of inducing variables and the properties of the logarithmic averaging operators. The main advantage of MR-IOWLAD operators is their design, which is specifically thought to aid in decision-making when a set of diverse regions with different properties must be considered. Moreover, the induced weighting vector and the distance measure mechanisms of the operator allow for the wider modeling of problems, including heterogeneous information and the complex attitudinal character of experts, when aiming for an ideal scenario. Along with analyzing the main properties of the IOWLAD operators, their families and specific cases, we also introduce some extensions, such as the induced generalized ordered weighted averaging IGOWLAD operator and Choquet integrals. We present the induced Choquet logarithmic distance averaging ICLD operator and the generalized induced Choquet logarithmic distance averaging IGCLD operator. Finally, an illustrative example is proposed, including real-world information retrieved from the United Nations World Statistics for global regions.