USMA Digital Commons (United States Military Academy, West Point)
Not a member yet
1355 research outputs found
Sort by
Extra Credit as a Spaced-Study Motivator
Full text linkA big challenge in academia is motivating students and helping them store information long-term. One known method to help students retain information is spaced learning, which allows for certain time intervals in which concepts are retested versus waiting until the end of a block to test a learning objective. One method to motivate students is implementing bonus opportunities that keep students engaged in the material. We combine these methods via a spaced daily versus bulk review program incentivized with bonus points. To investigate this method, we designed and implemented a classroom research experiment during Fall of 2022 at the United States Military Academy in three courses. We found that students who were incentivized to do either daily or bulk review bonus opportunities were primarily those who already were doing well in the class. We also found that those who did not pursue bonus opportunities, regardless of section, indicated that time was their limiting factor. We present our findings including quantitative and qualitative results via course grades, mid-semester and end-of-semester surveys, and anecdotal experiences with our students. We then discuss the merit and potential improvement of the experiment to better understand the relationship between student motivation and long-term information retention
Performance Analysis of Deep-Learning Based Open Set Recognition Algorithms for Network Intrusion Detection Systems
Full text linkOpen Set Recognition (OSR) is the ability of a machine learning (ML) algorithm to classify the known and recognize the unknown. In other words, OSR enables novelty detection in classification algorithms. This broader approach is critical to detect new types of attacks, including zero-days, thereby improving the effectiveness and efficiency of various ML-enabled mission-critical systems, such as cyber-physical, facial recognition, spam filtering, and cyber defense systems such as intrusion detection systems (IDS). In ML algorithms, like deep learning (DL) classifiers, hyperparameters control the learning process; their values affect other model parameters, such as weights and biases, which affect the performance of OSR algorithms. Moreover, OSR introduces additional parameters, making DL classifiers bigger and training them more computationally intensive. Determining the suitable set of hyperparameters and parameters is a computationally expensive task. Alternative OSR algorithms have demonstrated promising results on image datasets, but only limited studies have been performed in the context of IDS. This paper proposes OpenSetPerf, an empirical investigation of three prominent OSR algorithms using a current, real-world network intrusion detection systems (NIDS) benchmark dataset to discover the relationship between the DL-based OSR algorithm’s hyperparameter values and their performance. OpenSetperf evaluates these algorithms using quantitative studies with widely used ML performance evaluation metrics
Chapter 13 - Energy harvesting, charge, and mass transport considerations for reaction center-nanomaterial composites
No full textCombining proteins and abiotic substances such as electrodes and nanomaterials is as attractive as it is challenging. Photosynthetic reaction centers (RCs) convert light energy of a single photon to cause a charge separation resulting in an excited electron, typically with a high quantum yield, making them among the most attractive proteins to harness for energy capture. RCs from purple bacteria provide a unique combination of visible and near-infrared light utilization with long term stability. This chapter will consider not only RC structure and function, but also of RC density on the chosen electrode, the electrode material itself, as well as addressing overall efficiency and long-term stability. The overlap between multiple fields spanning biochemistry, biophysics, nano-material science, and chemical engineering inspired the authors to systematically review literature from these perspectives to highlight the interplay of photon energy reception, energy conversion, as well as electron and mass transport. Nanoparticles (NPs) similar in dimensions to natural antenna and RCs, that are all much shorter than the wavelength of light they interact with, provide a valuable insight into not only interfacing strategies, but also ease of synthesis and perhaps into RCs function, as well. Nanoparticle incorporation into plants is also reviewed. Multiple examples of common practices within the fields of electrochemistry, chemical preparation of biological, and nanomaterials, to include scale up challenges, are discussed. Past and current efforts in engineering of photovoltaic devices based on purple bacterial RCs are reviewed and compared based on current density in micro amps per square centimeter (mA cm2). Lastly, several opportunities are identified for the future of reaction center- electrode interfacial design
Hidden Stratagem - Microtargeting: The Future of Conflict
Full text linkIn September 2020, General Paul Nakasone, NSA Director and Commander of U.S. Cyber Command, called foreign influence operations “the next great disruptor.”[1] Nearly every intelligence agency in the United States government has been sounding the alarm over targeted influence operations enabled by social media companies since at least 2016, even though some of these operations started earlier. What often goes unstated and even less understood is the digital surveillance economy underlying these platforms and how this economic structure of trading free access for data collection about individuals’ lives poses a national security threat. Harvard sociologist Shoshana Zuboff calls this phenomenon “surveillance capitalism [which] unilaterally claims human experience as free raw material for translation into behavioral data.”[2] This behavioral data is transformed into increasingly accurate microtargeted advertising.[3] The new surveillance capitalism has enabled massive information warfare campaigns that can be aimed directly at target populations. The predictive power of surveillance capitalism is not only being leveraged for advertising success but increasingly harnessed for mass population control[4] enabled by massive amounts of individually identifiable, commercially available data with virtually no oversight or regulation.
This graphic novel is meant to illustrate the risks from microtargeting and the small, almost imperceptible actions that can shift the outcome of major events. Particularly in competition, the risks from commercial data collection and the weaponization of it towards adversary ends can shift momentum over a series of small movements. Supply chain disruptions can appear to be an accident, almost normal friction. Family members can be targeted to influence a key decision-maker. A long-term employee leaving for a better-paying job doesn’t render any alarms. A frustrated idealist can trigger a national security event because of what they’re consuming online – all individually targeted through algorithms enabled by commercial data collection. The threat from microtargeting isn’t a massive open wound. Instead, it’s death by a thousand cuts from an unseen knife.https://digitalcommons.usmalibrary.org/aci_books/1043/thumbnail.jp
Linear Algebra: Algorithms, Applications and Techniques
No full textLinear Algebra: Algorithms, Applications, and Techniques, Fourth Edition offers a modern and algorithmic approach to computation while providing clear and straightforward theoretical background information. The book guides readers through the major applications, with chapters on properties of real numbers, proof techniques, matrices, vector spaces, linear transformations, eigen values, and Euclidean inner products. Appendices on Jordan canonical forms and Markov chains are included for further study. This useful textbook presents broad and balanced views of theory, with key material highlighted and summarized in each chapter. To further support student practice, the book also includes ample exercises with answers and hints