Search NASA⌕ Search

SEARCH · Search NASA

Results for “Learning Cognitive”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 163 records · Page 9

Deep Learning for Unsupervised Insider Threat Detection in Structured Cybersecurity Data Streams

Analysis of an organization’s computer network activity is a key component of early detection and mitigation of insider threat, a growing concern for many organizations. Raw system logs are a prototypical example of streaming data that can quickly scale beyond the cognitive power of a human analyst. As a prospective filter for the human analyst, we present an online unsupervised deep learning approach to detect anomalous network activity from system logs in real time. Our models decompose anomaly scores into the contributions of individual user behavior features for increased interpretability to aid analysts reviewing potential cases of insider threat. Using the CERT Insider Threat Dataset v6.2 and threat detection recall as our performance metric, our novel deep and recurrent neural network models outperform Principal Component Analysis, Support Vector Machine and Isolation Forest based anomaly detection baselines. For our best model, the events labeled as insider threat activity in our dataset had an average anomaly score in the 95.53 percentile, demonstrating our approach’s potential to greatly reduce analyst workloads.

Tuor, Aaron R.↗

Information Systems for NASA's Aeronautics and Space Enterprises

The aerospace industry is being challenged to reduce costs and development time as well as utilize new technologies to improve product performance. Information technology (IT) is the key to providing revolutionary solutions to the challenges posed by the increasing complexity of NASA's aeronautics and space missions and the sophisticated nature of the systems that enable them. The NASA Ames vision is to develop technologies enabling the information age, expanding the frontiers of knowledge for aeronautics and space, improving America's competitive position, and inspiring future generations. Ames' missions to accomplish that vision include: 1) performing research to support the American aviation community through the unique integration of computation, experimentation, simulation and flight testing, 2) studying the health of our planet, understanding living systems in space and the origins of the universe, developing technologies for space flight, and 3) to research, develop and deliver information technologies and applications. Information technology may be defined as the use of advance computing systems to generate data, analyze data, transform data into knowledge and to use as an aid in the decision-making process. The knowledge from transformed data can be displayed in visual, virtual and multimedia environments. The decision-making process can be fully autonomous or aided by a cognitive processes, i.e., computational aids designed to leverage human capacities. IT Systems can learn as they go, developing the capability to make decisions or aid the decision making process on the basis of experiences gained using limited data inputs. In the future, information systems will be used to aid space mission synthesis, virtual aerospace system design, aid damaged aircraft during landing, perform robotic surgery, and monitor the health and status of spacecraft and planetary probes. NASA Ames through the Center of Excellence for Information Technology Office is leading the effort in pursuit of revolutionary, IT-based approaches to satisfying NASA's aeronautics and space requirements. The objective of the effort is to incorporate information technologies within each of the Agency's four Enterprises, i.e., Aeronautics and Space Transportation Technology, Earth, Science, Human Exploration and Development of Space and Space Sciences. The end results of these efforts for Enterprise programs and projects should be reduced cost, enhanced mission capability and expedited mission completion.

Kutler, Paul↗

Deep Learning for Unsupervised Insider Threat Detection in Structured Cyber Security Data Streams

Analysis of an organization's computer network activity is a key component of early detection and mitigation of insider threat, a growing concern for many organizations. Raw system logs are a prototypical example of streaming data that can quickly scale beyond the cognitive power of a human analyst. As a prospective filter for the human analyst, we present an online unsupervised deep learning approach to detect anomalous network activity from system logs in real time. Our models decompose anomaly scores into the contributions of individual user behavior features for increased interpretability to aid analysts reviewing potential cases of insider threat. Using the CERT Insider Threat Dataset v6.2 and threat detection recall as our performance metric, our novel deep and recurrent neural network models outperform Principal Component Analysis, Support Vector Machine and Isolation Forest based anomaly detection baselines. For our best model, the events labeled as insider threat activity in our dataset had an average anomaly score in the 95.53 percentile, demonstrating our approach's potential to greatly reduce analyst workloads.

Deep Learning, Insider Threat↗

Assuring Intelligent Systems: Contingency Management for UAS

Unmanned aircraft systems (UAS) collaborate with humans to operate in diverse, safety-critical applications. However, assurance technologies need to be integrated into the design process in order to guarantee safe behavior, thereby enabling UAS operations in the National Airspace System (NAS). In this paper, formal methods are integrated with learning-enabled systems representations. The generation and representation of knowledge are captured via monadic second-order logic rules in the cognitive architecture Soar. These rules are translated into timed automata, and a proof of correctness for the translation is provided so that safety and liveness properties can be checked in the formal verification environment Uppaal. This approach is agnostic to the learning mechanism used to generate the learned rules (e.g., chunking, etc.). An example of a fault-tolerant, learning-enabled UAS deciding which of four contingency procedures to execute under a lost link scenario while overflying an urban area is used to illustrate the approach.

Intelligent Systems↗

Structural Basis of Arc Binding to Synaptic Proteins: Implications for Cognitive Disease

Arc is a cellular immediate-early gene (IEG) that functions at excitatory synapses and is required for learning and memory. Here we report crystal structures of Arc subdomains that form a bi-lobar architecture remarkably similar to the capsid domain of human immunodeficiency virus (HIV) gag protein. Analysis indicates Arc originated from the Ty3/Gypsy retrotransposon family and was “domesticated” in higher vertebrates for synaptic functions. The Arc N-terminal lobe evolved a unique hydrophobic pocket that mediates intermolecular binding with synaptic proteins as resolved in complexes with TARPγ2 (Stargazin) and CaMKII peptides and is essential for Arc’s synaptic function. A consensus sequence for Arc binding identifies several additional partners that include genes implicated in schizophrenia. Arc N-lobe binding is inhibited by small chemicals suggesting Arc’s synaptic action may be druggable. Finally, these studies reveal the remarkable evolutionary origin of Arc and provide a structural basis for understanding Arc’s contribution to neural plasticity and disease.

59 BASIC BIOLOGICAL SCIENCES↗

Simulating Activities: Relating Motives, Deliberation and Attentive Coordination

Activities are located behaviors, taking time, conceived as socially meaningful, and usually involving interaction with tools and the environment. In modeling human cognition as a form of problem solving (goal-directed search and operator sequencing), cognitive science researchers have not adequately studied "off-task" activities (e.g., waiting), non-intellectual motives (e.g., hunger), sustaining a goal state (e.g., playful interaction), and coupled perceptual-motor dynamics (e.g., following someone). These aspects of human behavior have been considered in bits and pieces in past research, identified as scripts, human factors, behavior settings, ensemble, flow experience, and situated action. More broadly, activity theory provides a comprehensive framework relating motives, goals, and operations. This paper ties these ideas together, using examples from work life in a Canadian High Arctic research station. The emphasis is on simulating human behavior as it naturally occurs, such that "working" is understood as an aspect of living. The result is a synthesis of previously unrelated analytic perspectives and a broader appreciation of the nature of human cognition. Simulating activities in this comprehensive way is useful for understanding work practice, promoting learning, and designing better tools, including human-robot systems.

Clancey, William J.↗

The phylogenetic roots of addiction: compulsive drug seeking, natural and drug-sensitive reward, and the acquisition of learned habits

Our rational faculties permit us humans to maximize the utility of our actions. We perform a fundamental type of cost-benefit analysis in which we frame a problem, assign values to the different paths, and then choose from among a set of available options, the course that promises the most favorable outcome. So why then does addiction appear to be so impervious to the associated costs, so unaffected by undesired consequences, and ultimately so resistant to cognitive oversight? The answer may likely be found in the fact that the drivers for compulsive drug seeking and drug taking are located in affective brain circuits, circuits that are structured and patterned by learning with repeated activation. These deep processes exhibit significant resistance to control by our cognitive faculties. The general consensus is that addiction arises from mechanisms that overvalue the magnitude of reward, discount the associated risks, and thereby bias individuals towards compulsive pursuit of addictive drugs. Behavioral disruption and dependence appear to arise at the intersection of a number of connected but separate phenomena: expectations for the occurrence of specific events, behaviors that seek encounters with them, the ability to notice and learn nonrandom, co-occurring conditions, the prediction and valuation of consequences, the forming of enduring memories, and the drivers of focused behavior through compulsion, habits, and acquired routines. It is important to recognize that each one of these individual faculties are present and well developed across the entire phylogenetic tree of bilateral metazoans. The goal of this special volume is dedicated to exploring the degree to which inherent elements can account for addiction and addiction-associated phenomena. In much of the literature on addiction, the underlying processes are often viewed as distinctly mammalian, arising, in part, from the strong cognitive capacities of this taxon. Some phenomena may even be regarded to exist only in primates, or even solely in humans. This supposition arises from the fact that studies are conducted almost exclusively in mammals and primates, while evolutionary antecedents of the behavior are rarely considered. A more comprehensive perspective that examines drug reward and reinforcement in a wider range of organisms demonstrates that many of the component traits are actually well developed across the greater metazoan lineage, and they may well predate the emergence of a mammalian clade by a wide margin. The collection of papers assembled here supports the notion that the capacity to associate cues and quences has not arisen in mammals de novo. Rather, the neural mechanisms for detecting contingencies and for predicting future outcomes are very deeply rooted across broad phylogenetic divisions. Our understanding of an ability to associate paired events has been enriched by work in invertebrate preparations in both classical and operant conditioning scenarios [Cook and Carew, 1986, 1989a, 1989b]. The ability to learn allows us to connect cues and behavioral actions to their associated consequences. Pavlovian conditioning enriches surrounding cues with predictive value. Outcomes with positive valence generate appetitive responses and approach to the associated cues, while those perceived as aversive bring cue avoidance and withdrawal. Humans are not the only life forms capable of such short- and long-term modulations of behavior

59 BASIC BIOLOGICAL SCIENCES↗

Neural Development Under Conditions of Spaceflight

One of the key tasks the developing brain must learn is how to navigate within the environment. This skill depends on the brain's ability to establish memories of places and things in the environment so that it can form cognitive maps. Earth's gravity defines the plane of orientation of the spatial environment in which animals navigate, and cognitive maps are based on this plane of orientation. Given that experience during early development plays a key role in the development of other aspects of brain function, experience in a gravitational environment is likely to be essential for the proper organization of brain regions mediating learning and memory of spatial information. Since the hippocampus is the brain region responsible for cognitive mapping abilities, this study evaluated the development of hippocampal structure and function in rats that spent part of their early development in microgravity. Litters of male and female Sprague-Dawley rats were launched into space aboard the Space Shuttle Columbia on either postnatal day eight (P8) or 14 (P14) and remained in space for 16 days. Upon return to Earth, the rats were tested for their ability to remember spatial information and navigate using a variety of tests (the Morris water maze, a modified radial arm maze, and an open field apparatus). These rats were then tested physiologically to determine whether they exhibited normal synaptic plasticity in the hippocampus. In a separate group of rats (flight and controls), the hippocampus was analyzed using anatomical, molecular biological, and biochemical techniques immediately postlanding. There were remarkably few differences between the flight groups and their Earth-bound controls in either the navigation and spatial memory tasks or activity-induced synaptic plasticity. Microscopic and immunocytochemical analyses of the brain also did not reveal differences between flight animals and ground-based controls. These data suggest that, within the developmental window studied, microgravity has minimal long-term impact on cognitive mapping function and cellular substrates important for this function. Any differences due to development in microgravity were transient and returned to normal soon after return to Earth.

Kosik, Kenneth S.↗

Isoflurane induced cognitive impairment in aged rats through hippocampal calcineurin/NFAT signaling

Calcineurin (CaN) over-activation constrains synaptic plasticity and memory formation. Upon CaN activation, NFAT imports into the nucleus and guides its downstream genes, which also affect neuronal and synaptic function. Aberrant CaN/NFAT signaling involves in neurotoxicity and cognitive impairment in neurological disorders such as Alzheimer's disease, but its role in postoperative cognitive dysfunction (POCD) remains uninvestigated. Inhaled anesthetic isoflurane facilitates the development of POCD, and the present study investigated the role of CaN/NFAT signaling in isoflurane induced cognitive impairment of aged rats, and the therapeutic effects of CaN inhibitor cyclosporine A (CsA). The results indicated that hippocampal CaN activity increased and peaked at 6 h after isoflurane exposure, and NFAT, especially NFATc4, imported into the nucleus following CaN activation. Furthermore, phamacological inhibition of CaN by CsA markedly attenuated isoflurane induced aberrant CaN/NFATc4 signaling in the hippocampus, and rescued relevant spatial learning and memory impairment of aged rats. Overall, the study suggests hippocampal CaN/NFAT signaling as the upstream mechanism of isoflurane induced cognitive impairment, and provides potential therapeutic target and possible treatment methods for POCD. - Highlights: • Isoflurane induces hippocampal calcineurin activation. • Isoflurane induces hippocampal NFAT, especially NFATc4, nuclear import. • Cyclosporine A attenuates isoflurane induced aberrant calcineurin/NFAT signaling. • Cyclosporine A rescues isoflurane induced cognitive impairment. • Calcineurin/NFAT signaling is the upstream mechanism of isoflurane induced synaptic dysfunction and cognitive impairment.

60 APPLIED LIFE SCIENCES↗

Diagnosis and prognosis of Alzheimer's disease using brain morphometry and white matter connectomes

Accurate, reliable prediction of risk for Alzheimer's disease (AD) is essential for early, disease-modifying therapeutics. Multimodal MRI, such as structural and diffusion MRI, is likely to contain complementary information of neurodegenerative processes in AD. Here we tested the utility of the multimodal MRI (T1-weighted structure and diffusion MRI), combined with high-throughput brain phenotyping—morphometry and structural connectomics—and machine learning, as a diagnostic tool for AD. We used, firstly, a clinical cohort at a dementia clinic (National Health Insurance Service-Ilsan Hospital [NHIS-IH]; N = 211; 110 AD, 64 mild cognitive impairment [MCI], and 37 cognitively normal with subjective memory complaints [SMC]) to test the diagnostic models; and, secondly, Alzheimer's Disease Neuroimaging Initiative (ADNI)-2 to test the generalizability. Our machine learning models trained on the morphometric and connectome estimates (number of features = 34,646) showed optimal classification accuracy (AD/SMC: 97% accuracy, MCI/SMC: 83% accuracy; AD/MCI: 97% accuracy) in NHIS-IH cohort, outperforming a benchmark model (FLAIR-based white matter hyperintensity volumes). In ADNI-2 data, the combined connectome and morphometry model showed similar or superior accuracies (AD/HC: 96%; MCI/HC: 70%; AD/MCI: 75% accuracy) compared with the CSF biomarker model (t-tau, p-tau, and Amyloid β, and ratios). In predicting MCI to AD progression in a smaller cohort of ADNI-2 (n = 60), the morphometry model showed similar performance with 69% accuracy compared with CSF biomarker model with 70% accuracy. Our comparisons of the classifiers trained on structural MRI, diffusion MRI, FLAIR, and CSF biomarkers showed the promising utility of the white matter structural connectomes in classifying AD and MCI in addition to the widely used structural MRI-based morphometry, when combined with machine learning.

60 APPLIED LIFE SCIENCES↗

Effectiveness of Educational Intervention on the Congruence of Prostate and Rectal Contouring as Compared With a Gold Standard in Three-Dimensional Radiotherapy for Prostate

Purpose: To examine effects of a teaching intervention on precise delineation of the prostate and rectum during planning of three-dimensional conformal radiotherapy (3D-CRT) for prostate cancer. Methods and Materials: A pretest, posttest, randomized controlled group design was used. During pretest all participants contoured prostate and rectum on planning CT. Afterward, they participated in two types of workshops. The experimental group engaged in an interactive teaching session focused on prostate and rectum MR anatomy compared with CT anatomy. The control group focused on 3D-CRT planning without mention of prostate or rectal contouring. The experimental group practiced on fused MR-CT images, whereas the control group practiced on CT images. All participants completed the posttest. Results: Thirty-one trainees (12 male, 19 female) were randomly assigned to two groups, 17 in the experimental arm, and 14 in the control group. Seventeen felt familiar or very familiar with pelvic organ contouring, 12 somewhat, and 2 had never done it. Thirteen felt confident with organ contouring, 13 somewhat, and 5 not confident. The demographics and composition of groups were analyzed with chi{sup 2} and repeated-measures analysis of variance with the two groups (experimental or control) and two tests (pre- or posttest) as factors. Satisfaction with the course and long-term effects of the course on practice were assessed with immediate and delayed surveys. All performance variables showed a similar pattern of results. Conclusions: The training sessions improved the technical performance similarly in both groups. Participants were satisfied with the course content, and the delayed survey reflected that cognitively participants felt more confident with prostate and rectum contouring and would investigate opportunities to learn more about organ contouring.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Preserved number of entorhinal cortex layer II neurons in aged macaque monkeys

The perforant path, which consists of the projection from the layer II neurons of the entorhinal cortex to the outer molecular layer of the dentate gyrus, is a critical circuit involved in learning and memory formation. Accordingly, disturbances in this circuit may contribute to age-related cognitive deficits. In a previous study, we demonstrated a decrease in N-methyl-D-aspartate receptor subunit 1 immunofluorescence intensity in the outer molecular layer of aged macaque monkeys. In this study, we used the optical fractionator, a stereological method, to determine if a loss of layer II neurons occurred in the same animals in which the N-methyl-D-aspartate receptor subunit 1 alteration was observed. Our results revealed no significant differences in the number of layer II neurons between juvenile, young adult, and aged macaque monkeys. These results suggest that the circuit-specific decrease in N-methyl-D-aspartate receptor subunit 1 reported previously occurs in the absence of structural compromise of the perforant path, and thus may be linked to an age-related change in the physiological properties of this circuit.

Non-NASA Center↗

Learning the Task Management Space of an Aircraft Approach Model

Validating models of airspace operations is a particular challenge. These models are often aimed at finding and exploring safety violations, and aim to be accurate representations of real-world behavior. However, the rules governing the behavior are quite complex: nonlinear physics, operational modes, human behavior, and stochastic environmental concerns all determine the responses of the system. In this paper, we present a study on aircraft runway approaches as modeled in Georgia Tech's Work Models that Compute (WMC) simulation. We use a new learner, Genetic-Active Learning for Search-Based Software Engineering (GALE) to discover the Pareto frontiers defined by cognitive structures. These cognitive structures organize the prioritization and assignment of tasks of each pilot during approaches. We discuss the benefits of our approach, and also discuss future work necessary to enable uncertainty quantification.

Validation↗

The Hopper: A Wearable Robotic Device Testbed for Micro-Gravity Bone-Loading Proof-of-Concept

Wearable robotic systems are showing increased potential for addressing crew countermeasures needs. Wearable robots offer a compactness, programmability, and eccentric loading capability not present in more conventional exercise equipment. Correspondingly, advancements in the man to machine interface has progressed, allowing for higher loads to be applied directly to the person in new and novel ways. Recently, the X1 exoskeleton, a lower extremity wearable robot originally designed for mobility assistance and rehabilitation, underwent human subject testing to assess its potential as a knee dynamometer. This was of interest to NASA physiologists because currently strength is not assessed in flight due to hardware limitations, and thus there is a poor understanding of the time course of in-flight changes to muscle strength. The study concluded that the X1 compared well with the Biodex, the "gold standard" in terrestrial dynamometry, with coefficients of variation less than 6.0%. In a following study, the X1 powered ankle was evaluated for its efficacy in exercising calf muscles. Current on-orbit countermeasures equipment does not adequately protect the calf from atrophy. The results of this study were also positive (targeted muscle activity demonstrated via comparing pre- and post-exercise magnetic resonance imaging T2 measurements), again showing the efficacy of wearable robotic devices for addressing the countermeasure needs of our astronauts. Based on these successes and lessons learned, the Grasshopper was co-developed between IHMC (Florida Institute for Human and Machine Cognition) and NASA. The Grasshopper, or the Hopper for short, is a wearable robotic device designed to address muscle and bone density loss for astronauts spending extended periods of time in micro-gravity. The Grasshopper connects to the user's torso like a hiking backpack, over the shoulders and around the waist. At the feet are footplates that strap to the user. There are two actuators, one at each "knee" joint, which are capable of high fidelity torque control. Because the Hopper uses motors instead of gravity to create the load on the user, the device is suited for use on space missions. Exercise in zero-gravity conditions is critical to maintain muscle strength and bone mass. In operation, the actuators try to fold up, or collapse, the device, putting a compressive load between the user's feet and torso. This force is similar to carrying a heavy backpack. The user then bends and extends his or her knees, replicating a weightlifting squat exercise. The applied load is precisely controlled by a computer, and can be programmed to simulate gravitation loads or any desired load prescription, such as free-weight squat exercise. It is even possible to perform eccentric exercises, or negatives, without the need for a spotter. Because the hip joints, as well as the spine and long leg bones, are in the applied load path, there is the potential to stimulate bone growth, countering the typical bone loss when astronauts return from extended duration space travel.

Beck, C. E.↗

A Distributed Approach to High-Rate Delay Tolerant Networking Within a Virtualized Environment

The High-Rate Delay Tolerant Networking (HDTN) project has taken a distributed service-based approach to the development of a highly efficient delay tolerant networking (DTN) implementation. Through the analysis of many DTN implementations, system and mission requirements as well as the DTN protocol specifications, HDTN has worked to infuse modern computing technologies into the NASA approach to interplanetary networking. The initial use case of the HDTN software runs on a hypervisor representative of the International Space Station (ISS) DTN Gateway. In this scenario, multiple emulated payloads will send science data through HDTN to a mission operations center. HDTN will provide store and forward capability as well as network flow management. This paper discusses the infusion path of cognitive networking technologies in the NASA SCaN networks using the DTN architecture and protocols as the basis for cognitive routing and network management capabilities. HDTN has been developing the Bundle Protocol encoding and decoding mechanisms and messaging framework that can be used as the basis for integrating DTN with various learning and decision-making processes. The concepts of distributed computing, network virtualization, software defined networking and delay tolerant networking are basic building blocks which will further the development of cognitive networking. In addition to discussion of the HDTN software development and testing, this paper examines the role that each of these technologies play in the evolution of the current state of space networking into an intelligent network of networks.

Delay Tolerant Networking↗

A Distributed Approach to High-Rate Delay Tolerant Networking Within a Virtualized Environment

The High-Rate Delay Tolerant Networking (HDTN) project has taken a distributed service-based approach to the development of a highly efficient delay tolerant networking (DTN) implementation. Through the analysis of many DTN implementations, system and mission requirements as well as the DTN protocol specifications, HDTN has worked to infuse modern computing technologies into the NASA approach to interplanetary networking. The initial use case of the HDTN software runs on a hypervisor representative of the International Space Station (ISS) DTN Gateway. In this scenario, multiple emulated payloads will send science data through HDTN to a mission operations center. HDTN will provide store and forward capability as well as network flow management. This paper discusses the infusion path of cognitive networking technologies in the NASA Space Communications and Navigation (SCaN) networks using the DTN architecture and protocols as the basis for cognitive routing and network management capabilities. HDTN has been developing the Bundle Protocol encoding and decoding mechanisms and messaging framework that can be used as the basis for integrating DTN with various learning and decision-making processes. The concepts of distributed computing, network virtualization, software defined networking and delay tolerant networking are basic building blocks which will further the development of cognitive networking. In addition to discussion of the HDTN software development and testing, this paper examines the role that each of these technologies play in the evolution of the current state of space networking into an intelligent network of networks.

Rachel Mary Dudukovich↗

Toward an embedded training tool for Deep Space Network operations

There are three issues to consider when building an embedded training system for a task domain involving the operation of complex equipment: (1) how skill is acquired in the task domain; (2) how the training system should be designed to assist in the acquisition of the skill, and more specifically, how an intelligent tutor could aid in learning; and (3) whether it is feasible to incorporate the resulting training system into the operational environment. This paper describes how these issues have been addressed in a prototype training system that was developed for operations in NASA's Deep Space Network (DSN). The first two issues were addressed by building an executable cognitive model of problem solving and skill acquisition of the task domain and then using the model to design an intelligent tutor. The cognitive model was developed in Soar for the DSN's Link Monitor and Control (LMC) system; it led to several insights about learning in the task domain that were used to design an intelligent tutor called REACT that implements a method called 'impasse-driven tutoring'. REACT is one component of the LMC training system, which also includes a communications link simulator and a graphical user interface. A pilot study of the LMC training system indicates that REACT shows promise as an effective way for helping operators to quickly acquire expert skills.

Hill, Randall W., Jr.↗