https://axon.cs.byu.edu/~martinez/classes/678/Presentations/Yao.pdf

Extreme learning machine: Theory and applications

I have developed "tennis elbow" from lugging this book around the past four weeks, but it is worth the pain, the effort, and the aspirin. It is also worth the (relatively speaking) bargain price. Including appendixes, this book contains 894 pages of text. The entire panorama of the neural sciences is surveyed and examined, and it is comprehensive in its scope, from genomes to social behaviors. The editors explicitly state that the book is designed as "an introductory text for students of biology, behavior, and medicine," but it is hard to imagine any audience, interested in any fragment of neuroscience at any level of sophistication, that would not enjoy this book. The editors have done a masterful job of weaving together the biologic, the behavioral, and the clinical sciences into a single tapestry in which everyone from the molecular biologist to the practicing psychiatrist can find and appreciate his or

Principles of Neural Science

This paper surveys the field of reinforcement learning from a computer-science perspective. It is written to be accessible to researchers familiar with machine learning. Both the historical basis of the field and a broad selection of current work are summarized. Reinforcement learning is the problem faced by an agent that learns behavior through trial-and-error interactions with a dynamic environment. The work described here has a resemblance to work in psychology, but differs considerably in the details and in the use of the word "reinforcement." The paper discusses central issues of reinforcement learning, including trading off exploration and exploitation, establishing the foundations of the field via Markov decision theory, learning from delayed reinforcement, constructing empirical models to accelerate learning, making use of generalization and hierarchy, and coping with hidden state. It concludes with a survey of some implemented systems and an assessment of the practical utility of current methods for reinforcement learning.

/pdf/reinforcement-learning-a-survey-38lmyylg2y.pdf

Reinforcement learning: a survey

Planning algorithms are impacting technical disciplines and industries around the world, including robotics, computer-aided design, manufacturing, computer graphics, aerospace applications, drug design, and protein folding. This coherent and comprehensive book unifies material from several sources, including robotics, control theory, artificial intelligence, and algorithms. The treatment is centered on robot motion planning but integrates material on planning in discrete spaces. A major part of the book is devoted to planning under uncertainty, including decision theory, Markov decision processes, and information spaces, which are the “configuration spaces” of all sensor-based planning problems. The last part of the book delves into planning under differential constraints that arise when automating the motions of virtually any mechanical system. Developed from courses taught by the author, the book is intended for students, engineers, and researchers in robotics, artificial intelligence, and control theory as well as computer graphics, algorithms, and computational biology.

Planning Algorithms: Introductory Material

This paper surveys the field of reinforcement learning from a computer-science perspective. It is written to be accessible to researchers familiar with machine learning. Both the historical basis of the field and a broad selection of current work are summarized. Reinforcement learning is the problem faced by an agent that learns behavior through trial-and-error interactions with a dynamic environment. The work described here has a resemblance to work in psychology, but differs considerably in the details and in the use of the word ``reinforcement.'' The paper discusses central issues of reinforcement learning, including trading off exploration and exploitation, establishing the foundations of the field via Markov decision theory, learning from delayed reinforcement, constructing empirical models to accelerate learning, making use of generalization and hierarchy, and coping with hidden state. It concludes with a survey of some implemented systems and an assessment of the practical utility of current methods for reinforcement learning.

Reinforcement Learning: A Survey

Optical communications for transport aircraft are discussed. The problem involves: increasing demand for radio-frequency bands from an enlarging pool of users (aircraft, ground and sea vehicles, fleet operators, traffic control centers, and commercial radio and television); desirability of providing high-bandwidth dedicated communications to and from every aircraft in the National Airspace System; need to support communications, navigation, and surveillance for a growing number of aircraft; and improved meteorological observations by use of probe aircraft. The solution involves: optical signal transmission support very high data rates; optical transmission of signals between aircraft, orbiting satellites, and ground stations, where unobstructed line-of-sight is available; conventional radio transmissions of signals between aircraft and ground stations, where optical line-of-sight is unavailable; and radio priority given to aircraft in weather.

/pdf/optical-communications-for-transport-aircraft-16woc6l504.pdf

Optical communications for transport aircraft

Much remains to be learned about the hazards of low altitude wind shear to aviation. New research should be conducted on the nature of the atmospheric environment, on aircraft performance, and on guidance and control aids. In conducting this research, it is important to distinguish between near-term and far-term objectives, between basic and applied research, and between uses of results for aircraft design or for real-time implementation. Advances in on-board electronics can be applied to assuring that aircraft of all classes have near optimal protection against wind shear hazards.

Unresolved issues in wind shear encounters

A computerized intelligent cockpit aid can increase flight-crew awareness of wind shear, improving avoidance decisions. The FAA Windshear Training Aid (WTA) provides guidelines for assessing the risk of wind shear encounter from meteorological evidence. Use of these guidelines in the cockpit is complicated by uncertainty surrounding meteorological knowledge of wind shear. Bayesian network representation is discussed as a means of modeling this uncertain knowledge in a computer. A probabilistic model of the WTA guidelines using Bayesian network representation is presented. This model combines evidence from sources of varying reliability and incorporates results from meteorological studies of wind shear. The probabilistic model can provide flight crews with meaningful estimates of risk to aid their decisions, using evidence from a variety of sources and a base of meteorological knowledge.

Probabilistic reasoning for intelligent wind shear avoidance

Severe weight penalties can occur if the sizes of Orbital Sortie Vehicle stages are not near their optimum values. In particular, single-stage rocket configurations with or without “zero-stage” boosters are necessarily much larger than two-stage configurations. Conventional cryogenic propellant combinations provide a reasonable range of vehicle sizes. By comparison, one storable-liquid propellant combination would require vehicle size to be excessive, while two “exotic” propellant combinations do not appear to provide enough size reduction to warrant their use. Second-stage structural efficiency has a major effect on overall vehicle size. It appears feasible to design a two-stage Orbital Sortie Vehicle whose gross lift-off weight is 50–70 times the payload weight.

Propulsion and staging considerations for an orbital sortie vehicle

This paper explores closed-loop digital control of a high-speed impact printer hammer in the presence of sensor measurement noise and disturbances. Conventional control of high-speed impact printers is accomplished without feedback; hence disturbances, such as mechanical and magnetic interaction between hammers and hammer settle-out period, force long delays between hammer firings to assure proper time and velocity of impact. By computer simulation, we show the feasibility of using the solenoid actuator to control both time and velocity of impact

Robert F. Stengel

Papers

Optical communications for transport aircraft

Unresolved issues in wind shear encounters

Probabilistic reasoning for intelligent wind shear avoidance

Propulsion and staging considerations for an orbital sortie vehicle

Closed-loop digital control of an impact printer hammer.