Results 241 to 250 of about 946,507 (289)
Some of the next articles are maybe not open access.
Principled reward shaping for reinforcement learning via lyapunov stability theory
Neurocomputing, 2020Abstract Reinforcement learning (RL) suffers from the designation in reward function and the large computational iterating steps until convergence. How to accelerate the training process in RL plays a vital role. In this paper, we proposed a Lyapunov function based approach to shape the reward function which can effectively accelerate the training ...
Yunlong Dong, Xiuchuan Tang, Ye Yuan
openaire +1 more source
Stability Certificates for Neural Network Learning-based Controllers using Robust Control Theory
2021 American Control Conference (ACC), 2021Providing stability guarantees for controllers that use neural networks can be challenging. Robust control theoretic tools are used to derive a framework for providing nominal stability guarantees – stability guarantees for a known nominal system – controlled by a learning-based neural network controller.
Hoang Hai Nguyen +5 more
openaire +1 more source
STABILITY RESULTS IN LEARNING THEORY
Analysis and Applications, 2005The problem of proving generalization bounds for the performance of learning algorithms can be formulated as a problem of bounding the bias and variance of estimators of the expected error. We show how various stability assumptions can be employed for this purpose.
Rakhlin, Alexander +2 more
openaire +2 more sources
Stability theory of universal learning network
1996 IEEE International Conference on Systems, Man and Cybernetics. Information Intelligence and Systems (Cat. No.96CH35929), 2002Higher order derivatives of the universal learning network (ULN) has been previously derived by forward and backward propagation computing methods, which can model and control the large scale complicated systems such as industrial plants, economic, social and life phenomena. In this paper, a new concept of nth order asymptotic orbital stability for the
K. Hirasawa +3 more
openaire +1 more source
A NOTE ON STABILITY OF ERROR BOUNDS IN STATISTICAL LEARNING THEORY
Analysis and Applications, 2011We consider a wide class of error bounds developed in the context of statistical learning theory which are expressed in terms of functionals of the regression function, for instance, its norm in a reproducing kernel Hilbert space or other functional space.
Li, Ming, Caponnetto, Andrea
openaire +2 more sources
Experimentally verified Iterative Learning Control based on repetitive process stability theory
2012 American Control Conference (ACC), 2012This paper gives new results on the design and experimental evaluation of an Iterative Learning Control (ILC) law in a repetitive process setting. The experimental results given are from a gantry robot facility that has been extensively used in the benchmarking of linear model based ILC designs.
P. Dabkowski +7 more
openaire +1 more source
Analysis of robust control using stability theory of universal learning networks
IEEE SMC'99 Conference Proceedings. 1999 IEEE International Conference on Systems, Man, and Cybernetics (Cat. No.99CH37028), 2003Nth order asymptotic orbital stability analysis method has been proposed to determine whether a nonlinear system is stable or not with large fluctuations of the system states. In this paper, we discuss the stability of robust control of a nonlinear crane system using this method.
null Yunqing Yu +3 more
openaire +1 more source
The Journal of Chemical Physics, 2018
We use a combination of machine learning techniques and high-throughput density-functional theory calculations to explore ternary compounds with the AB2C2 composition. We chose the two most common intermetallic prototypes for this composition, namely, the tI10-CeAl2Ga2 and the tP10-FeMo2B2 structures.
Jonathan Schmidt +3 more
openaire +2 more sources
We use a combination of machine learning techniques and high-throughput density-functional theory calculations to explore ternary compounds with the AB2C2 composition. We chose the two most common intermetallic prototypes for this composition, namely, the tI10-CeAl2Ga2 and the tP10-FeMo2B2 structures.
Jonathan Schmidt +3 more
openaire +2 more sources
IFAC Proceedings Volumes, 2013
Abstract Iterative learning control has been developed for systems that repeat the same task over a finite duration with resetting to the starting location once each repetition, or trial, is complete. The novel feature is the use of information generated on the previous trial to compute the control input for the next one and the basic problem is to ...
Wojciech Paszke +2 more
openaire +1 more source
Abstract Iterative learning control has been developed for systems that repeat the same task over a finite duration with resetting to the starting location once each repetition, or trial, is complete. The novel feature is the use of information generated on the previous trial to compute the control input for the next one and the basic problem is to ...
Wojciech Paszke +2 more
openaire +1 more source