Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Learning Theory”

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 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
NASA NTRS2011

Modeling Humans as Reinforcement Learners: How to Predict Human Behavior in Multi-Stage Games

This paper introduces a novel framework for modeling interacting humans in a multi-stage game environment by combining concepts from game theory and reinforcement learning. The proposed model has the following desirable characteristics: (1) Bounded rational players, (2) strategic (i.e., players account for one anothers reward functions), and (3) is computationally feasible even on moderately large real-world systems. To do this we extend level-K reasoning to policy space to, for the first time, be able to handle multiple time steps. This allows us to decompose the problem into a series of smaller ones where we can apply standard reinforcement learning algorithms. We investigate these ideas in a cyber-battle scenario over a smart power grid and discuss the relationship between the behavior predicted by our model and what one might expect of real human defenders and attackers.

Lee, Ritchie↗
NASA NTRS1993

Indirect decentralized repetitive control

Learning control refers to controllers that learn to improve their performance at executing a given task, based on experience performing this specific task. In a previous work, the authors presented a theory of indirect decentralized learning control based on use of indirect adaptive control concepts employing simultaneous identification and control. This paper extends these results to apply to the indirect repetitive control problem in which a periodic (i.e., repetitive) command is given to a control system. Decentralized indirect repetitive control algorithms are presented that have guaranteed convergence to zero tracking error under very general conditions. The original motivation of the repetitive control and learning control fields was learning in robots doing repetitive tasks such as on an assembly line. This paper starts with decentralized discrete time systems, and progresses to the robot application, modeling the robot as a time varying linear system in the neighborhood of the desired trajectory. Decentralized repetitive control is natural for this application because the feedback control for link rotations is normally implemented in a decentralized manner, treating each link as if it is independent of the other links.

Lee, Soo Cheol↗
NASA NTRS1979

Aircraft adaptive learning control

The optimal control theory of stochastic linear systems is discussed in terms of the advantages of distributed-control systems, and the control of randomly-sampled systems. An optimal solution to longitudinal control is derived and applied to the F-8 DFBW aircraft. A randomly-sampled linear process model with additive process and noise is developed.

Lee, P. S. T.↗
NASA NTRS1990

System identification in the repetition domain

Procedures for system identification using realization theory in conjunction with learning control ideas are developed. The Markov parameters of the system are identified by combining data from repeated experiments. Three approaches are discussed for identification of as many Markov parameters as sample points in the experiment. Making use of all the parameters, realization theory is then employed to determine the system order and to obtain a minimal order representation. The first two approaches are non-recursive, which in the case of noise-free data yields a one step solution. The third approach uses a recursive formulation rendered from adaptive control but modified for successive experiments. A simple example shows the numerical convergence of the identified parameters as a function of the number of experiments. The procedure presented herein is an extension of the existing Eigensystem Realization Algorithm (ERA), which has been successfully applied for system identification of large structures.

Juang, Jer-Nan↗
NASA NTRS2004

Adaptive, Distributed Control of Constrained Multi-Agent Systems

Product Distribution (PO) theory was recently developed as a broad framework for analyzing and optimizing distributed systems. Here we demonstrate its use for adaptive distributed control of Multi-Agent Systems (MASS), i.e., for distributed stochastic optimization using MAS s. First we review one motivation of PD theory, as the information-theoretic extension of conventional full-rationality game theory to the case of bounded rational agents. In this extension the equilibrium of the game is the optimizer of a Lagrangian of the (Probability dist&&on on the joint state of the agents. When the game in question is a team game with constraints, that equilibrium optimizes the expected value of the team game utility, subject to those constraints. One common way to find that equilibrium is to have each agent run a Reinforcement Learning (E) algorithm. PD theory reveals this to be a particular type of search algorithm for minimizing the Lagrangian. Typically that algorithm i s quite inefficient. A more principled alternative is to use a variant of Newton's method to minimize the Lagrangian. Here we compare this alternative to RL-based search in three sets of computer experiments. These are the N Queen s problem and bin-packing problem from the optimization literature, and the Bar problem from the distributed RL literature. Our results confirm that the PD-theory-based approach outperforms the RL-based scheme in all three domains.

Bieniawski, Stefan↗
NASA NTRS2010

Multiple Kernel Learning for Heterogeneous Anomaly Detection: Algorithm and Aviation Safety Case Study

The world-wide aviation system is one of the most complex dynamical systems ever developed and is generating data at an extremely rapid rate. Most modern commercial aircraft record several hundred flight parameters including information from the guidance, navigation, and control systems, the avionics and propulsion systems, and the pilot inputs into the aircraft. These parameters may be continuous measurements or binary or categorical measurements recorded in one second intervals for the duration of the flight. Currently, most approaches to aviation safety are reactive, meaning that they are designed to react to an aviation safety incident or accident. In this paper, we discuss a novel approach based on the theory of multiple kernel learning to detect potential safety anomalies in very large data bases of discrete and continuous data from world-wide operations of commercial fleets. We pose a general anomaly detection problem which includes both discrete and continuous data streams, where we assume that the discrete streams have a causal influence on the continuous streams. We also assume that atypical sequence of events in the discrete streams can lead to off-nominal system performance. We discuss the application domain, novel algorithms, and also discuss results on real-world data sets. Our algorithm uncovers operationally significant events in high dimensional data streams in the aviation industry which are not detectable using state of the art methods

Das, Santanu↗
NASA NTRS1992

Artificial Intelligence Research Branch future plans

This report contains information on the activities of the Artificial Intelligence Research Branch (FIA) at NASA Ames Research Center (ARC) in 1992, as well as planned work in 1993. These activities span a range from basic scientific research through engineering development to fielded NASA applications, particularly those applications that are enabled by basic research carried out in FIA. Work is conducted in-house and through collaborative partners in academia and industry. All of our work has research themes with a dual commitment to technical excellence and applicability to NASA short, medium, and long-term problems. FIA acts as the Agency's lead organization for research aspects of artificial intelligence, working closely with a second research laboratory at the Jet Propulsion Laboratory (JPL) and AI applications groups throughout all NASA centers. This report is organized along three major research themes: (1) Planning and Scheduling: deciding on a sequence of actions to achieve a set of complex goals and determining when to execute those actions and how to allocate resources to carry them out; (2) Machine Learning: techniques for forming theories about natural and man-made phenomena; and for improving the problem-solving performance of computational systems over time; and (3) Research on the acquisition, representation, and utilization of knowledge in support of diagnosis design of engineered systems and analysis of actual systems.

Stewart, Helen↗
NASA NTRS2016

Theoretical Aeroacoustics: Compiled Mathematical Derivations of Fereidoun 'Feri' Farassat

Dr. Fereidoun 'Feri' Farassat was a theoretical aero-acoustician at the National Aeronautics and Space Administration (NASA) Langley Research Center. This book contains technical derivations, notes, and classes that Dr. Farassat produced during his professional career. The layout of the book has been carefully crafted so that foundational ideas through advanced theories, which altered the technical discipline of aeroacoustics, build upon one another. The book can be used to understand the theories of acoustics and learn one contemporary aeroacoustic prediction approach made popular by Dr. Farassat. Most importantly, this book gives the general reader insight into how one of NASA's best aeroacoustics theoreticians thought, constructed, and solved problems throughout his career.

Miller, Steven A. E.↗
NASA NTRS2005

List Models of Procedure Learning

This paper presents a new theory of the initial stages of skill acquisition and then employs the theory to model current and future training programs for fight management systems (FMSs) in modern commercial airliners like the Boeing 777 and the Airbus A320. The theoretical foundations for the theory are a new synthesis of the literature on human memory and the latest version of the ACT-R theory of skill acquisition.

Matessa, Michael P.↗
NASA NTRS2008

Simulated Students and Classroom Use of Model-Based Intelligent Tutoring

Two educational uses of models and simulations: 1) Students create models and use simulations ; and 2) Researchers create models of learners to guide development of reliably effective materials. Cognitive tutors simulate and support tutoring - data is crucial to create effective model. Pittsburgh Science of Learning Center: Resources for modeling, authoring, experimentation. Repository of data and theory. Examples of advanced modeling efforts: SimStudent learns rule-based model. Help-seeking model: Tutors metacognition. Scooter uses machine learning detectors of student engagement.

Koedinger, Kenneth R.↗
NASA NTRS1990

Learning control for minimizing a quadratic cost during repetitions of a task

In many applications, control systems are asked to perform the same task repeatedly. Learning control laws have been developed over the last few years that allow the controller to improve its performance each repetition, and to converge to zero error in tracking a desired trajectory. This paper generates a new type of learning control law that learns to minimize a quadratic cost function for tracking. Besides being of interest in its own right, this objective alleviates the need to specify a desired trajectory that can actually be performed by the system. The approach used here is to adapt appropriate methods from numerical optimization theory in order to produce learning control algorithms that adjust the system command from repetition to repetition in order to converge to the quadratic cost optimal trajectory.

Longman, Richard W.↗
NASA NTRS2004

Product Distributions for Distributed Optimization

With connections to bounded rational game theory, information theory and statistical mechanics, Product Distribution (PD) theory provides a new framework for performing distributed optimization. Furthermore, PD theory extends and formalizes Collective Intelligence, thus connecting distributed optimization to distributed Reinforcement Learning (FU). This paper provides an overview of PD theory and details an algorithm for performing optimization derived from it. The approach is demonstrated on two unconstrained optimization problems, one with discrete variables and one with continuous variables. To highlight the connections between PD theory and distributed FU, the results are compared with those obtained using distributed reinforcement learning inspired optimization approaches. The inter-relationship of the techniques is discussed.

Bieniawski, Stefan R.↗
NASA NTRS1997

Goldstone-Apple Valley Radio Telescope System Theory of Operation

The purpose of this learning module is to enable learners to describe how the Goldstone-Apple Valley Radio Telescope (GAVRT) system functions in support of Apple Valley Science and Technology Center's (AVSTC) client schools' radio astronomy activities.

Radio Telescope↗
NASA NTRS1992

Tradeoffs in the utility of learned knowledge

Planning systems which make use of domain theories can produce more accurate plans and achieve more goals as the quality of their domain knowledge improves. MTR, a multi-strategy learning system, was designed to learn from system failures and improve domain knowledge used in planning. However, augmented domain knowledge can decrease planning efficiency. We describe how improved knowledge that becomes expensive to use can be approximated to yield calculated tradeoffs in accuracy and efficiency.

Kedar, Smadar↗