Search

Your search keyword '"Rajasekaran, A."' showing total 1,169 results
Start Over Author "Rajasekaran, A." Publication Year Range This year
1,169 results on '"Rajasekaran, A."'

1. Insulin and leptin oscillations license food-entrained browning and metabolic flexibility

Author

Mattar, Pamela, Reginato, Andressa, Lavados, Christian, Das, Debajyoti, Kalyani, Manu, Martinez-Lopez, Nuria, Sharma, Mridul, Skovbjerg, Grethe, Skytte, Jacob Lercke, Roostalu, Urmas, Subbarayan, Rajasekaran, Picarda, Elodie, Zang, Xingxing, Zhang, Jinghang, Guha, Chandan, Schwartz, Gary, Rajbhandari, Prashant, and Singh, Rajat

Subjects
Biochemistry and Cell Biology, Biological Sciences, Diabetes, Obesity, Nutrition, 2.1 Biological and endogenous factors, Metabolic and endocrine, Animals, Leptin, Insulin, Adipose Tissue, Brown, Mice, Mice, Inbred C57BL, Energy Metabolism, Adipose Tissue, White, Male, Feeding Behavior, CP: Metabolism, ILC2, browning, circadian, insulin, intermittent fasting, leptin, oscillations, time-restricted feeding, Medical Physiology, Biological sciences
Abstract

Timed feeding drives adipose browning, although the integrative mechanisms for the same remain unclear. Here, we show that twice-a-night (TAN) feeding generates biphasic oscillations of circulating insulin and leptin, representing their entrainment by timed feeding. Insulin and leptin surges lead to marked cellular, functional, and metabolic remodeling of subcutaneous white adipose tissue (sWAT), resulting in increased energy expenditure. Single-cell RNA-sequencing (scRNA-seq) analyses and flow cytometry demonstrate a role for insulin and leptin surges in innate lymphoid type 2 (ILC2) cell recruitment and sWAT browning, since sWAT depot denervation or loss of leptin or insulin receptor signaling or ILC2 recruitment each dampens TAN feeding-induced sWAT remodeling and energy expenditure. Consistently, recreating insulin and leptin oscillations via once-a-day timed co-injections is sufficient to favorably remodel innervated sWAT. Innervation is necessary for sWAT remodeling, since denervation of sWAT, but not brown adipose tissue (BAT), blocks TAN-induced sWAT remodeling and resolution of inflammation. In sum, reorganization of nutrient-sensitive pathways remodels sWAT and drives the metabolic benefits of timed feeding.

Published
2024

2. Linguistic isolation correlates with length of stay and mortality for pediatric oncology patients in California.

Author

Ennett, Samuel, Das, Akansha, Burcham, Megan, Fitzgerald, Robert, Boville, Brian, Rajasekaran, Surender, Kortz, Teresa, and Leimanis-Laurens, Mara

Subjects
cancer, health disparities, health inequities, linguistic isolation, medical length of stay, mortality, pediatric intensive care unit, pediatrics, Humans, California, Length of Stay, Child, Female, Neoplasms, Male, Intensive Care Units, Pediatric, Child, Preschool, Adolescent, Infant, Hospital Mortality
Abstract

OBJECTIVE: To evaluate social drivers of health and how they impact pediatric oncology patients clinical outcomes during pediatric intensive care unit (PICU) admission via correlation with patient ZIP codes. METHODS: Demographic, clinical, and outcome variables from Virtual Pediatric Systems®, LLC for oncology patients (2009-2021) in California PICUs (excluding postoperative) using 3-digit ZIP Codes with social drivers of health variables linguistic isolation, poverty, race/ethnicity, and education abstracted from American Community Survey data for 3-digit ZIP Codes using the Environmental Protection Agencys EJScreen tool. Outcomes of length of stay (LOS), mortality, acuity scores, were compared with social variables. RESULTS: Positive correlation between mortality and minority racial groups (Hispanic/Latino) across ZIP Codes (correlation coefficients of 0.45 (95% CI: 0.22-0.64, p

Published
2024

3. APEER: Automatic Prompt Engineering Enhances Large Language Model Reranking

Author

Jin, Can, Peng, Hongwu, Zhao, Shiyu, Wang, Zhenting, Xu, Wujiang, Han, Ligong, Zhao, Jiahui, Zhong, Kai, Rajasekaran, Sanguthevar, and Metaxas, Dimitris N.

Subjects
Computer Science - Artificial Intelligence
Abstract

Large Language Models (LLMs) have significantly enhanced Information Retrieval (IR) across various modules, such as reranking. Despite impressive performance, current zero-shot relevance ranking with LLMs heavily relies on human prompt engineering. Existing automatic prompt engineering algorithms primarily focus on language modeling and classification tasks, leaving the domain of IR, particularly reranking, underexplored. Directly applying current prompt engineering algorithms to relevance ranking is challenging due to the integration of query and long passage pairs in the input, where the ranking complexity surpasses classification tasks. To reduce human effort and unlock the potential of prompt optimization in reranking, we introduce a novel automatic prompt engineering algorithm named APEER. APEER iteratively generates refined prompts through feedback and preference optimization. Extensive experiments with four LLMs and ten datasets demonstrate the substantial performance improvement of APEER over existing state-of-the-art (SoTA) manual prompts. Furthermore, we find that the prompts generated by APEER exhibit better transferability across diverse tasks and LLMs. Code is available at https://github.com/jincan333/APEER.

Published
2024

4. Fluctuations of Eigenvalues for Generalized Patterned Gaussian Random Matrices

Author

Rajasekaran, Frederick

Subjects
Mathematics - Probability, 60B20, 60F05, 15B52
Abstract

In this work, we study a class of random matrices which interpolate between the Wigner matrix model and various types of patterned random matrices such as random Toeplitz, Hankel, and circulant matrices. The interpolation mechanism is through the correlations of the entries, and thus these interpolating models are highly inhomogeneous in their correlation structure. Historically, the study of random matrices has focused on homogeneous models, i.e., those with imposed structure (such as independence of the entries), as such restrictions significantly simplify computations related to the models. However, in this paper we demonstrate that for these interpolating inhomogenous models, the fluctuations of the linear eigenvalue statistics are approximately Gaussian. To handle the difficulties that come with inhomogeneity in the entries, we incorporate combinatorial arguments and recent tools from non-asymptotic random matrix theory., Comment: 22 pages

Published
2024

32. MLTCP: Congestion Control for DNN Training

Author

Rajasekaran, Sudarsanan, Narang, Sanjoli, Zabreyko, Anton A., and Ghobadi, Manya

Subjects
Computer Science - Networking and Internet Architecture, Computer Science - Distributed, Parallel, and Cluster Computing, Computer Science - Machine Learning
Abstract

We present MLTCP, a technique to augment today's congestion control algorithms to accelerate DNN training jobs in shared GPU clusters. MLTCP enables the communication phases of jobs that compete for network bandwidth to interleave with each other, thereby utilizing the network efficiently. At the heart of MLTCP lies a very simple principle based on a key conceptual insight: DNN training flows should scale their congestion window size based on the number of bytes sent at each training iteration. We show that integrating this principle into today's congestion control protocols is straightforward: by adding 30-60 lines of code to Reno, CUBIC, or DCQCN, MLTCP stabilizes flows of different jobs into an interleaved state within a few training iterations, regardless of the number of competing flows or the start time of each flow. Our experiments with popular DNN training jobs demonstrate that enabling MLTCP accelerates the average and 99th percentile training iteration time by up to 2x and 4x, respectively.

Published
2024

Catalog

Books, media, physical & digital resources
See catalog results