skip to main content


Title: Unraveling Metabolic and Proteomic Features in Soybean Plants in Response to Copper Hydroxide Nanowires Compared to a Commercial Fertilizer
Award ID(s):
1901515
NSF-PAR ID:
10325238
Author(s) / Creator(s):
; ; ; ;
Date Published:
Journal Name:
Environmental Science & Technology
Volume:
55
Issue:
20
ISSN:
0013-936X
Page Range / eLocation ID:
13477 to 13489
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Learning to optimize (L2O) has gained increasing popularity, which automates the design of optimizers by data-driven approaches. However, current L2O methods often suffer from poor generalization performance in at least two folds: (i) applying the L2O-learned optimizer to unseen optimizees, in terms of lowering their loss function values (optimizer generalization, or “generalizable learning of optimizers”); and (ii) the test performance of an optimizee (itself as a machine learning model), trained by the optimizer, in terms of the accuracy over unseen data (optimizee generalization, or “learning to generalize”). While the optimizer generalization has been recently studied, the optimizee generalization (or learning to generalize) has not been rigorously studied in the L2O context, which is the aim of this paper. We first theoretically establish an implicit connection between the local entropy and the Hessian, and hence unify their roles in the handcrafted design of generalizable optimizers as equivalent metrics of the landscape flatness of loss functions. We then propose to incorporate these two metrics as flatness-aware regularizers into the L2O framework in order to meta-train optimizers to learn to generalize, and theoretically show that such generalization ability can be learned during the L2O meta-training process and then transformed to the optimizee loss function. Extensive experiments consistently validate the effectiveness of our proposals with substantially improved generalization on multiple sophisticated L2O models and diverse optimizees. 
    more » « less
  2. null (Ed.)
  3. Marriage and divorce expectations predict family life and personal outcomes. Understanding how expectations are associated with varying characteristics over emerging adulthood (ages 18–28 years) will inform understanding of emerging adult development. Panel Study of Income Dynamics Transition into Adulthood 2005–2015 data were used in hybrid-effects ordinal logistic regression to parse interindividual and intraindividual variation associated with relationship experiences, socioeconomic and contextual characteristics, and mental and emotional well-being. Partnerships were associated with optimistic expectations: Both dating and cohabiting predicted greater marriage expectations and lower divorce expectations within individuals. Between individuals, greater time in full-time employment predicted more positive marital expectations, greater responsibility was associated with lower marital expectations, stronger religious identity predicted higher marital expectations and lower divorce expectations, having been arrested predicted greater divorce expectations, greater well-being predicted greater marriage expectations, and older age predicted lower marriage expectations. Both between and within individuals, greater worry predicted lower marriage expectations.

     
    more » « less
  4. null (Ed.)