An official website of the United States government
Using tags within a mark-recapture framework allows researchers to assess population size and connectivity. Such methods have been applied in coastal zone habitats to monitor salt marsh restoration success by comparing the movement patterns of Mummichogs (Fundulus heteroclitus) between restored and natural marshes. Visible Implant Elastomer (VIE) tags are commonly used to tag small fish like Mummichogs, though the retention and survival of small fish using this method varies between studies, producing uncertainty during mark-recapture-based approaches. To address this, we conducted a laboratory experiment to determine the rate of tag loss and mortality of VIE tags on Mummichogs of two size classes (greater or less than 61 mm) and across different taggers. Tag loss and mortality increased over time, and the latter significantly varied between taggers. We then developed a predictive model, R package ‘retmort’, to account for the effect of this increase on mark-recapture studies. When adapted to a series of published works, our model provided rational estimates of tagging error for multiple species and tagging methods. Of the case studies the model was applied to (n = 26), 15 resulted in a percent standard error greater than 5%, signaling a significant percent of error due to uncounted, tagged animals. By not accounting for these individuals, recapture studies, particularly those that assess restoration efforts and coastal resilience, could underestimate the effects of those projects, leading to superfluous restoration efforts and erroneous recapture data for species with low tag retention and high mortality rates.
more »
« less
Perturbation confusion in forward automatic differentiation of higher-order functions
Abstract Automatic differentiation (AD) is a technique for augmenting computer programs to compute derivatives. The essence of AD in its forward accumulation mode is to attach perturbations to each number, and propagate these through the computation by overloading the arithmetic operators. When derivatives are nested, the distinct derivative calculations, and their associated perturbations, must be distinguished. This is typically accomplished by creating a unique tag for each derivative calculation and tagging the perturbations. We exhibit a subtle bug, present in fielded implementations which support derivatives of higher-order functions, in which perturbations are confused despite the tagging machinery, leading to incorrect results. The essence of the bug is as follows: a unique tag is needed for each derivative calculation, but in existing implementations unique tags are created when taking the derivative of a function at a point. When taking derivatives of higher-order functions, these need not correspond! We exhibit a simple example: a higher-order function f whose derivative at a point x , namely f ′( x ), is itself a function which calculates a derivative. This situation arises naturally when taking derivatives of curried functions. Two potential solutions are presented, and their deficiencies discussed. One uses eta expansion to delay the creation of fresh tags in order to put them into one-to-one correspondence with derivative calculations. The other wraps outputs of derivative operators with tag substitution machinery. Both solutions seem very difficult to implement without violating the desirable complexity guarantees of forward AD.
more »
« less
- PAR ID:
- 10177535
- Date Published:
- Journal Name:
- Journal of Functional Programming
- Volume:
- 29
- ISSN:
- 0956-7968
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
We introduce a deep architecture named HoD-Net to enable high-order differentiability for deep learning. HoD-Net is based on and generalizes the complex-step finite difference (CSFD) method. While similar to classic finite difference, CSFD approaches the derivative of a function from a higher-dimension complex domain, leading to highly accurate and robust differentiation computation without numerical stability issues. This method can be coupled with backpropagation and adjoint perturbation methods for an efficient calculation of high-order derivatives. We show how this numerical scheme can be leveraged in challenging deep learning problems, such as high-order network training, deep learning-based physics simulation, and neural differential equations.more » « less
-
Abstract Honey bees are vital pollinators and can be used to monitor the landscape. Consequently, interest in mounting technologies onto bees to track foraging behaviors is increasing. The barrier to entry is steep, in part because the methodology for fastening tags to bees, and the success rates, are often missing from publications. We tested six factors suspected to influence the presence and tag retention rates of nurse honey bees after their introduction to hives, and followed bees until foraging age. We also compared reintroducing foragers to their maternal colony using the best method for nurse bees to releasing them in front of their maternal hive and allowing them to fly back unaided. Nurses were most likely to be present in the hive with their tag still attached when introduced using an introduction cage at night. Glue type was important, but may further be influenced by tag material. Foragers were most likely to be present with a tag attached if released in front of their colony. Preparation and introduction techniques influence the likelihood of tagged honey bee survival and of the tags remaining attached, which should be considered when executing honey bee tagging and tracking experiments.more » « less
-
Due to the exponential growth of scientific publications on the Web, there is a pressing need to tag each paper with fine-grained topics so that researchers can track their interested fields of study rather than drowning in the whole literature. Scientific literature tagging is beyond a pure multi-label text classification task because papers on the Web are prevalently accompanied by metadata information such as venues, authors, and references, which may serve as additional signals to infer relevant tags. Although there have been studies making use of metadata in academic paper classification, their focus is often restricted to one or two scientific fields (e.g., computer science and biomedicine) and to one specific model. In this work, we systematically study the effect of metadata on scientific literature tagging across 19 fields. We select three representative multi-label classifiers (i.e., a bag-of-words model, a sequence-based model, and a pre-trained language model) and explore their performance change in scientific literature tagging when metadata are fed to the classifiers as additional features. We observe some ubiquitous patterns of metadata’s effects across all fields (e.g., venues are consistently beneficial to paper tagging in almost all cases), as well as some unique patterns in fields other than computer science and biomedicine, which are not explored in previous studies.more » « less
-
Azide moieties, unique linear species containing three nitrogen atoms, represent an attractive class of molecular tag for hyperpolarized magnetic resonance imaging (HP-MRI). Here we demonstrate ( 15 N) 3 -azide-containing molecules exhibit long-lasting hyperpolarization lifetimes up to 9.8 min at 1 T with remarkably high polarization levels up to 11.6% in water, thus establishing ( 15 N) 3 -azide as a powerful spin storage for hyperpolarization. A single ( 15 N)-labeled azide has also been examined as an effective alternative tag with long-lived hyperpolarization. A variety of biologically important molecules are studied in this work, including choline, glucose, amino acid, and drug derivatives, demonstrating great potential of 15 N-labeled azides as universal hyperpolarized tags for nuclear magnetic resonance imaging applications.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1017/S095679681900008X
