Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “learning rate”

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 19 records
All sourcesNASA NTRSDOE OSTI
NASA NTRS1992

Automatic learning rate adjustment for self-supervising autonomous robot control

Described is an application in which an Artificial Neural Network (ANN) controls the positioning of a robot arm with five degrees of freedom by using visual feedback provided by two cameras. This application and the specific ANN model, local liner maps, are based on the work of Ritter, Martinetz, and Schulten. We extended their approach by generating a filtered, average positioning error from the continuous camera feedback and by coupling the learning rate to this error. When the network learns to position the arm, the positioning error decreases and so does the learning rate until the system stabilizes at a minimum error and learning rate. This abolishes the need for a predetermined cooling schedule. The automatic cooling procedure results in a closed loop control with no distinction between a learning phase and a production phase. If the positioning error suddenly starts to increase due to an internal failure such as a broken joint, or an environmental change such as a camera moving, the learning rate increases accordingly. Thus, learning is automatically activated and the network adapts to the new condition after which the error decreases again and learning is 'shut off'. The automatic cooling is therefore a prerequisite for the autonomy and the fault tolerance of the system.

Arras, Michael K.↗
NASA NTRS2005

Method of Real-Time Principal-Component Analysis

Dominant-element-based gradient descent and dynamic initial learning rate (DOGEDYN) is a method of sequential principal-component analysis (PCA) that is well suited for such applications as data compression and extraction of features from sets of data. In comparison with a prior method of gradient-descent-based sequential PCA, this method offers a greater rate of learning convergence. Like the prior method, DOGEDYN can be implemented in software. However, the main advantage of DOGEDYN over the prior method lies in the facts that it requires less computation and can be implemented in simpler hardware. It should be possible to implement DOGEDYN in compact, low-power, very-large-scale integrated (VLSI) circuitry that could process data in real time.

Duong, Tuan↗
NASA NTRS2005

Real-Time Principal-Component Analysis

A recently written computer program implements dominant-element-based gradient descent and dynamic initial learning rate (DOGEDYN), which was described in Method of Real-Time Principal-Component Analysis (NPO-40034) NASA Tech Briefs, Vol. 29, No. 1 (January 2005), page 59. To recapitulate: DOGEDYN is a method of sequential principal-component analysis (PCA) suitable for such applications as data compression and extraction of features from sets of data. In DOGEDYN, input data are represented as a sequence of vectors acquired at sampling times. The learning algorithm in DOGEDYN involves sequential extraction of principal vectors by means of a gradient descent in which only the dominant element is used at each iteration. Each iteration includes updating of elements of a weight matrix by amounts proportional to a dynamic initial learning rate chosen to increase the rate of convergence by compensating for the energy lost through the previous extraction of principal components. In comparison with a prior method of gradient-descent-based sequential PCA, DOGEDYN involves less computation and offers a greater rate of learning convergence. The sequential DOGEDYN computations require less memory than would parallel computations for the same purpose. The DOGEDYN software can be executed on a personal computer.

Duong, Vu↗
NASA NTRS1995

Locomotion training of legged robots using hybrid machine learning techniques

In this study artificial neural networks and fuzzy logic are used to control the jumping behavior of a three-link uniped robot. The biped locomotion control problem is an increment of the uniped locomotion control. Study of legged locomotion dynamics indicates that a hierarchical controller is required to control the behavior of a legged robot. A structured control strategy is suggested which includes navigator, motion planner, biped coordinator and uniped controllers. A three-link uniped robot simulation is developed to be used as the plant. Neurocontrollers were trained both online and offline. In the case of on-line training, a reinforcement learning technique was used to train the neurocontroller to make the robot jump to a specified height. After several hundred iterations of training, the plant output achieved an accuracy of 7.4%. However, when jump distance and body angular momentum were also included in the control objectives, training time became impractically long. In the case of off-line training, a three-layered backpropagation (BP) network was first used with three inputs, three outputs and 15 to 40 hidden nodes. Pre-generated data were presented to the network with a learning rate as low as 0.003 in order to reach convergence. The low learning rate required for convergence resulted in a very slow training process which took weeks to learn 460 examples. After training, performance of the neurocontroller was rather poor. Consequently, the BP network was replaced by a Cerebeller Model Articulation Controller (CMAC) network. Subsequent experiments described in this document show that the CMAC network is more suitable to the solution of uniped locomotion control problems in terms of both learning efficiency and performance. A new approach is introduced in this report, viz., a self-organizing multiagent cerebeller model for fuzzy-neural control of uniped locomotion is suggested to improve training efficiency. This is currently being evaluated for a possible patent by NASA, Johnson Space Center. An alternative modular approach is also developed which uses separate controllers for each stage of the running stride. A self-organizing fuzzy-neural controller controls the height, distance and angular momentum of the stride. A CMAC-based controller controls the movement of the leg from the time the foot leaves the ground to the time of landing. Because the leg joints are controlled at each time step during flight, movement is smooth and obstacles can be avoided. Initial results indicate that this approach can yield fast, accurate results.

Simon, William E.↗
NASA NTRS2006

Active Learning with Irrelevant Examples

Active learning algorithms attempt to accelerate the learning process by requesting labels for the most informative items first. In real-world problems, however, there may exist unlabeled items that are irrelevant to the user's classification goals. Queries about these points slow down learning because they provide no information about the problem of interest. We have observed that when irrelevant items are present, active learning can perform worse than random selection, requiring more time (queries) to achieve the same level of accuracy. Therefore, we propose a novel approach, Relevance Bias, in which the active learner combines its default selection heuristic with the output of a simultaneously trained relevance classifier to favor items that are likely to be both informative and relevant. In our experiments on a real-world problem and two benchmark datasets, the Relevance Bias approach significantly improved the learning rate of three different active learning approaches.

machine learning↗
NASA NTRS2002

The Relationship Between Fidelity and Learning in Aviation Training and Assessment

Flight simulators can be designed to train pilots or assess their flight performance. Low-Fidelity simulators maximize the initial learning rate of novice pilots and minimize initial costs; whereas, expensive, high-fidelity simulators predict the realworld in-flight performance of expert pilots (Fink & Shriver, 1978 Hays & Singer 1989; Kinkade & Wheaton. 1972). Although intuitively appealing and intellectually convenient to generalize concepts of learning and assessment, what holds true for the role of fidelity in assessment may not always hold true for learning, and vice versa. To bring clarity to this issue, the author distinguishes the role of fidelity in learning from its role in assessment as a function of skill level by applying the hypothesis of Alessi (1988) and reviewing the Laughery, Ditzian, and Houtman (1982) study on simulator validity. Alessi hypothesized that there is it point beyond which one additional unit of flight-simulator fidelity results in a diminished rate of learning. The author of this current paper also suggests the existence of an optimal point beyond which one additional unit of flight-simulator fidelity results in a diminished rate of practical assessment of nonexpert pilot performance.

Noble, Cliff↗
NASA NTRS2005

Development of Advanced Verification and Validation Procedures and Tools for the Certification of Learning Systems in Aerospace Applications

Adaptive control technologies that incorporate learning algorithms have been proposed to enable automatic flight control and vehicle recovery, autonomous flight, and to maintain vehicle performance in the face of unknown, changing, or poorly defined operating environments. In order for adaptive control systems to be used in safety-critical aerospace applications, they must be proven to be highly safe and reliable. Rigorous methods for adaptive software verification and validation must be developed to ensure that control system software failures will not occur. Of central importance in this regard is the need to establish reliable methods that guarantee convergent learning, rapid convergence (learning) rate, and algorithm stability. This paper presents the major problems of adaptive control systems that use learning to improve performance. The paper then presents the major procedures and tools presently developed or currently being developed to enable the verification, validation, and ultimate certification of these adaptive control systems. These technologies include the application of automated program analysis methods, techniques to improve the learning process, analytical methods to verify stability, methods to automatically synthesize code, simulation and test methods, and tools to provide on-line software assurance.

Jacklin, Stephen↗
NASA NTRS1994

Fuzzy self-learning control for magnetic servo system

It is known that an effective control system is the key condition for successful implementation of high-performance magnetic servo systems. Major issues to design such control systems are nonlinearity; unmodeled dynamics, such as secondary effects for copper resistance, stray fields, and saturation; and that disturbance rejection for the load effect reacts directly on the servo system without transmission elements. One typical approach to design control systems under these conditions is a special type of nonlinear feedback called gain scheduling. It accommodates linear regulators whose parameters are changed as a function of operating conditions in a preprogrammed way. In this paper, an on-line learning fuzzy control strategy is proposed. To inherit the wealth of linear control design, the relations between linear feedback and fuzzy logic controllers have been established. The exercise of engineering axioms of linear control design is thus transformed into tuning of appropriate fuzzy parameters. Furthermore, fuzzy logic control brings the domain of candidate control laws from linear into nonlinear, and brings new prospects into design of the local controllers. On the other hand, a self-learning scheme is utilized to automatically tune the fuzzy rule base. It is based on network learning infrastructure; statistical approximation to assign credit; animal learning method to update the reinforcement map with a fast learning rate; and temporal difference predictive scheme to optimize the control laws. Different from supervised and statistical unsupervised learning schemes, the proposed method learns on-line from past experience and information from the process and forms a rule base of an FLC system from randomly assigned initial control rules.

Tarn, J. H.↗
NASA NTRS2018

Automated Pneumothorax Diagnosis using Deep Neural Networks

Thoracic ultrasound can provide information leading to rapid diagnosis of pneumothorax with improved accuracy over the standard physical examination and with higher sensitivity than anteroposterior chest radiography. However, the clinical We have Furthermore, remote environments, such as the battlefield or deep-space exploration, may lack expertise for diagnosing developed an automated image interpretation pipeline for the analysis of thoracic ultrasound data and the classification of pneumothorax events to provide decision support in such situations. Our pipeline consists of image preprocessing, data augmentation, and deep learning architectures for medical diagnosis. In this work, we demonstrate that robust, accurate interpretation of chest images and video can be achieved using deep neural networks. A number of novel image processing techniques were employed to achieve this result. Affine transformations were applied for data augmentation. Hyperparameters were optimized for learning rate, dropout regularization, batch size, and epoch iteration by a sequential model-based Bayesian approach. In addition, we utilized pretrained architecturesinterpretation of a patient medical image is highly operator dependent. certain pathologies., applying transfer learning and fine-tuning techniques to fully connected layers. Our pipeline yielded binary classification validation accuracies of 98.3% for M-mode images and 99.8% with B-mode video frames.

US Army collaboration↗
NASA NTRS2009

Active Learning with Irrelevant Examples

An improved active learning method has been devised for training data classifiers. One example of a data classifier is the algorithm used by the United States Postal Service since the 1960s to recognize scans of handwritten digits for processing zip codes. Active learning algorithms enable rapid training with minimal investment of time on the part of human experts to provide training examples consisting of correctly classified (labeled) input data. They function by identifying which examples would be most profitable for a human expert to label. The goal is to maximize classifier accuracy while minimizing the number of examples the expert must label. Although there are several well-established methods for active learning, they may not operate well when irrelevant examples are present in the data set. That is, they may select an item for labeling that the expert simply cannot assign to any of the valid classes. In the context of classifying handwritten digits, the irrelevant items may include stray marks, smudges, and mis-scans. Querying the expert about these items results in wasted time or erroneous labels, if the expert is forced to assign the item to one of the valid classes. In contrast, the new algorithm provides a specific mechanism for avoiding querying the irrelevant items. This algorithm has two components: an active learner (which could be a conventional active learning algorithm) and a relevance classifier. The combination of these components yields a method, denoted Relevance Bias, that enables the active learner to avoid querying irrelevant data so as to increase its learning rate and efficiency when irrelevant items are present. The algorithm collects irrelevant data in a set of rejected examples, then trains the relevance classifier to distinguish between labeled (relevant) training examples and the rejected ones. The active learner combines its ranking of the items with the probability that they are relevant to yield a final decision about which item to present to the expert for labeling. Experiments on several data sets have demonstrated that the Relevance Bias approach significantly decreases the number of irrelevant items queried and also accelerates learning speed.

Wagstaff, Kiri↗
NASA NTRS1979

Space shuttle solid rocket booster cost-per-flight analysis technique

A cost per flight computer model is described which considers: traffic model, component attrition, hardware useful life, turnaround time for refurbishment, manufacturing rates, learning curves on the time to perform tasks, cost improvement curves on quantity hardware buys, inflation, spares philosophy, long lead, hardware funding requirements, and other logistics and scheduling constraints. Additional uses of the model include assessing the cost per flight impact of changing major space shuttle program parameters and searching for opportunities to make cost effective management decisions.

Forney, J. A.↗
NASA NTRS1990

Satellite servicing economic study

Previous studies have shown that satellite servicing is cost effective; however, all of these studies were of different formats, dollar year, learning rates, availability, etc. Therefore, it was difficult to correlate any useful trends from these studies. The reviewed study was initiated to correlate the economic data into a common data base, using a common set of assumptions. A selected set of existed funded programs was then analyzed to provide an independent analysis of the servicing options and potential economic benefits.

Source record↗
NASA NTRS1990

Satellite servicing economic study

Previous studies have shown that satellite servicing is cost effective; however, all of these studies were of different formats, dollar year, learning rates, availability, etc. Threfore, it was difficult to correlate any useful trends from these studies. The reviewed study was initiated to correlate the economic data into a common data base, using a common set of assumptions. A selected set of existed funded programs was then analyzed to provide an independent analysis of the servicing options and potential economic benefits.

Source record↗
NASA NTRS1990

Learning reliable manipulation strategies without initial physical models

A description is given of a robot, possessing limited sensory and effectory capabilities but no initial model of the effects of its actions on the world, that acquires such a model through exploration, practice, and observation. By acquiring an increasingly correct model of its actions, it generates increasingly successful plans to achieve its goals. In an apparently nondeterministic world, achieving reliability requires the identification of reliable actions and a preference for using such actions. Furthermore, by selecting its training actions carefully, the robot can significantly improve its learning rate.

Christiansen, Alan D.↗
NASA NTRS1993

Learning characteristics of a space-time neural network as a tether skiprope observer

The Software Technology Laboratory at the Johnson Space Center is testing a Space Time Neural Network (STNN) for observing tether oscillations present during retrieval of a tethered satellite. Proper identification of tether oscillations, known as 'skiprope' motion, is vital to safe retrieval of the tethered satellite. Our studies indicate that STNN has certain learning characteristics that must be understood properly to utilize this type of neural network for the tethered satellite problem. We present our findings on the learning characteristics including a learning rate versus momentum performance table.

Lea, Robert N.↗
NASA NTRS1993

Two generalizations of Kohonen clustering

The relationship between the sequential hard c-means (SHCM), learning vector quantization (LVQ), and fuzzy c-means (FCM) clustering algorithms is discussed. LVQ and SHCM suffer from several major problems. For example, they depend heavily on initialization. If the initial values of the cluster centers are outside the convex hull of the input data, such algorithms, even if they terminate, may not produce meaningful results in terms of prototypes for cluster representation. This is due in part to the fact that they update only the winning prototype for every input vector. The impact and interaction of these two families with Kohonen's self-organizing feature mapping (SOFM), which is not a clustering method, but which often leads ideas to clustering algorithms is discussed. Then two generalizations of LVQ that are explicitly designed as clustering algorithms are presented; these algorithms are referred to as generalized LVQ = GLVQ; and fuzzy LVQ = FLVQ. Learning rules are derived to optimize an objective function whose goal is to produce 'good clusters'. GLVQ/FLVQ (may) update every node in the clustering net for each input vector. Neither GLVQ nor FLVQ depends upon a choice for the update neighborhood or learning rate distribution - these are taken care of automatically. Segmentation of a gray tone image is used as a typical application of these algorithms to illustrate the performance of GLVQ/FLVQ.

Bezdek, James C.↗
NASA NTRS1992

Image segmentation using fuzzy LVQ clustering networks

In this note we formulate image segmentation as a clustering problem. Feature vectors extracted from a raw image are clustered into subregions, thereby segmenting the image. A fuzzy generalization of a Kohonen learning vector quantization (LVQ) which integrates the Fuzzy c-Means (FCM) model with the learning rate and updating strategies of the LVQ is used for this task. This network, which segments images in an unsupervised manner, is thus related to the FCM optimization problem. Numerical examples on photographic and magnetic resonance images are given to illustrate this approach to image segmentation.

Tsao, Eric Chen-Kuo↗
NASA NTRS1992

Learning characteristics of a space-time neural network as a tether skiprope observer

The Software Technology Laboratory at JSC is testing a Space Time Neural Network (STNN) for observing tether oscillations present during retrieval of a tethered satellite. Proper identification of tether oscillations, known as 'skiprope' motion, is vital to safe retrieval of the tethered satellite. Our studies indicate that STNN has certain learning characteristics that must be understood properly to utilize this type of neural network for the tethered satellite problem. We present our findings on the learning characteristics including a learning rate versus momentum performance table.

Lea, Robert N.↗