An official website of the United States government
In this paper, we explore the potential of utilizing time-stamps as labels for Deep Learning from webcams, surveillance cameras, and other fixed viewpoint image situations. Specifically, we explore if learning to classify images by the time they were taken uncovers interesting patterns and behaviors in the scenes captured by these cameras. We describe approaches to building datasets with large quantities of images and their accompanying labels, making them suitable for large-scale deep learning approaches. We share our results from the initial deep learning experiments.
more »
« less
This content will become publicly available on July 3, 2026
It’s about Time: Conceptions of Time in Decision Making about Elementary Science Improvement
- Award ID(s):
- 1761057
- PAR ID:
- 10615108
- Publisher / Repository:
- The University of Chicago Press
- Date Published:
- Journal Name:
- The Elementary School Journal
- ISSN:
- 0013-5984
- Page Range / eLocation ID:
- 000 to 000
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Recurrent Neural Networks (RNNs) are important tools for processing sequential data such as time-series or video. Interpretability is defined as the ability to be understood by a person and is different from explainability, which is the ability to be explained in a mathematical formulation. A key interpretability issue with RNNs is that it is not clear how each hidden state per time step contributes to the decision-making process in a quantitative manner. We propose NeuroView-RNN as a family of new RNN architectures that explains how all the time steps are used for the decision-making process. Each member of the family is derived from a standard RNN architecture by concatenation of the hidden steps into a global linear classifier. The global linear classifier has all the hidden states as the input, so the weights of the classifier have a linear mapping to the hidden states. Hence, from the weights, NeuroView-RNN can quantify how important each time step is to a particular decision. As a bonus, NeuroView-RNN also offers higher accuracy in many cases compared to the RNNs and their variants. We showcase the benefits of NeuroView-RNN by evaluating on a multitude of diverse time-series datasets.more » « less
-
Rutgers University is a R1 Research Institution with $929M in Sponsored Research funding in FY 2023 and $747M in Research Expenditures in 2022 as per the HERD survey. Since forming in 2019, the Data, Analytics and Business Intelligence team within the Office for Research has grown to three members, all with the mission of providing accurate and timely data about the Rutgers research enterprise for both internal leadership and decision makers across the university. The Office for Research leadership was receiving complaints from the field about sponsored research awards taking too long to be set up. The Data, Analytics and Business Intelligence team was tasked to find out: 1. Is this a common issue? 2. Is this a localized issue? 3. What’s causing this? Our team determined that by calculating the start and end of processes, and by categorizing responsible party within those processes, we can figure out what an average user might experience and then dig deeper into the outliers. We involved IT to extract the data from the award set up system into the data warehouse, and ultimately a Tableau data source. We also met with the business team to better understand the data in their system and define how their processes should be categorized, and if any of their processes might need to change to help generate important data. The final deliverable was a published Tableau Dashboard that leadership and the business unit could access that shows turnaround time trends broken down in various ways, with the option to drill down into outliers without leaving the dashboard. Since then, we’ve been able to replicate this method across other Office for Research business units including the teams managing IRB and IACUC protocols. Our presentation would include: 1. The problem that was presented to us 2. The process we followed 3. Samples of the dashboards we created 4. The outcome and lessons learned 5. Guiding questions on how other schools can use this method and replicate it across multiple systems within their Research Officesmore » « less
