Interactive technology has a complicated relationship with recreation in nature. Many people have praised, and many have lamented the impact of interactive technology on recreation in nature. Because nature recreation has important wellness benefits and interactive technology is likely to remain a part of nature recreation, there is a need to design interactive technology for nature recreation. Unfortunately, little generalized knowledge exists on how to design such technology. We create new intermediate design knowledge for interactive technology in nature recreation by drawing from others’ work, our prior work, and specifically Borgmann and Verbeek’s philosophies of technology. Our contribution is a framework based on a decomposition of engagement into nine facets related to engagement with place, time, and community. Four examples demonstrate the descriptive and generative power of the framework. This framework may enable the creation of interactive systems that complement rather than compete with nature recreation and may better preserve the wellness benefits of nature recreation.

We introduce an open-source software platform for developing microgrid planning tools. Our platform is composed of a computational layer developed in Python; a MySQL database layer; a REST API developed in Flask; a web app front end developed in Flask, HTML templates, and JavaScript; containerized deployment through Docker; and high-performance computing integration using Slurm. Our base capabilities include user accounts with authentication, user-defined distributed energy resource components and microgrids, user uploads of power load data, a core simulation method, and a microgrid sizing method. These capabilities are all integrated into a user-friendly web application, which is designed to be customized and extended. Whereas our platform already includes a useful set of microgrid planning tools, our vision is for it to bridge the active academic analytical modeling research for microgrid planning into deployable software tools that can be readily used by practitioners. In addition to describing our current capabilities, this paper details how our platform design facilitates easy adoption by other researchers developing analytical methods for microgrid planning. History: Accepted by Ted Ralphs, Area Editor for Software Tools. Funding: This research was supported by the NextSTEP Program, sponsored by the Office of Naval Research; by Naval Facilities Engineering Systems Command as part of the Navy Shore Energy Technology Transition and Integration program; and by the Director of Operational Energy, Deputy Assistant Secretary of the U.S. Navy. Supplemental Material: The software that supports the findings of this study is available within the paper and its Supplemental Information ( https://pubsonline.informs.org/doi/suppl/10.1287/ijoc.2023.0336 ) as well as from the IJOC GitHub software repository ( https://github.com/INFORMSJoC/2023.0336 ). The complete IJOC Software and Data Repository is available at https://informsjoc.github.io/ .

Visual Grounding (VG) methods in Visual Question Answering (VQA) attempt to improve VQA performance by strengthening a model's reliance on question-relevant visual information.The presence of such relevant information in the visual input is typically assumed in training and testing.This assumption, however, is inherently flawed when dealing with imperfect image representations common in large-scale VQA, where the information carried by visual features frequently deviates from expected ground-truth contents.As a result, training and testing of VG-methods is performed with largely inaccurate data, which obstructs proper assessment of their potential benefits.In this study, we demonstrate that current evaluation schemes for VG-methods are problematic due to the flawed assumption of availability of relevant visual information.Our experiments show that these methods can be much more effective when evaluation conditions are corrected.Code is provided on GitHub 1 .

In this work we present an online ASR system that is able to discriminate voice commands directed to an operationable screen from irrelevant speech segments. For classification of the sound segments we explored several features that are based on prosody as well as properties generated during the decoding process. For a vocabulary of 259 words and more than 10k possible commands, our realtime Verbal Command Detector managed to detect 88.3% of the commands in our evaluation data while maintaining a low False Positive Rate (FPR) of 1.5%. On an evaluation task using an episode of Star Trek, our system was able to detect 91.2% of all commands with a FPR of 1.8% with only minor adjustments. The system is part of and used in the Smart Control Room at the Fraunhofer IOSB in Karlsruhe [1], an experimental smart environment that uses multiple input modalities for crisis response.

We present an algorithm for preprocessing a class of stochastic shortest-path problems on networks that have no negative cost cycles, almost surely. Our method adds utility to existing frameworks by significantly reducing input problem sizes and thereby increasing computational tractability. Given random costs with finite lower and upper bounds on each edge, our algorithm removes edges that cannot be in any optimal solution to the deterministic shortest-path problem, for any realization of the random costs. Although this problem is NP-complete, our algorithm efficiently preprocesses nearly all edges in a given network. We provide computational results both on sparse networks from PSPLIB—a well-known project evaluation and review technique library [Kolisch, R., A. Sprecher. 1996. PSPLIB—A project scheduling problem library. Eur. J. Oper. Res. 96(1) 205–216]—and dense synthetic ones: on average, less than 0.1% of the edges in the PSPLIB instances and 0.5% of the edges in the dense instances remain unclassified after preprocessing.

Abstract In this article, we present a stochastic shortest path problem that we refer to as the Most Likely Path Problem (MLPP). We demonstrate that optimal solutions to the MLPP are not composed of optimal subpaths, which limits the computational tractability of exact solution methods. On series‐parallel networks, we produce analytical bounds for the MLPP's optimality indices, the probabilities of given paths in the network being shortest, and compute these bounds efficiently via numerical integration. These bounds can also be used independently of the MLPP to gain further understanding for paths of interest that are identified by other stochastic shortest path frameworks, e.g., robust shortest paths or expected shortest paths. Additionally, we present a heuristic method that uses dynamic programming and ordinal optimization to identify an MLP on series‐parallel networks. Our computational study shows our bounds to be tight in a majority of test networks and shows our heuristic to be both efficient and highly accurate for identifying an MLP in all test networks. © 2010 Wiley Periodicals, Inc. NETWORKS, Vol. 58(1), 68–80 2011