Relevant bibliographies by topics / Dynamic optimal learning rate

Academic literature on the topic 'Dynamic optimal learning rate'

Author: Grafiati
Published: 4 June 2021
Last updated: 20 February 2023

Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles

Select a source type:

Contents

Consult the lists of relevant articles, books, theses, conference reports, and other scholarly sources on the topic 'Dynamic optimal learning rate.'

Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.

You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.

Journal articles on the topic "Dynamic optimal learning rate"

1

Chinrungrueng, C., and C. H. Sequin. "Optimal adaptive k-means algorithm with dynamic adjustment of learning rate." IEEE Transactions on Neural Networks 6, no. 1 (1995): 157–69. http://dx.doi.org/10.1109/72.363440.

Full text
APA, Harvard, Vancouver, ISO, and other styles
2

Zhu, Yingqiu, Danyang Huang, Yuan Gao, Rui Wu, Yu Chen, Bo Zhang, and Hansheng Wang. "Automatic, dynamic, and nearly optimal learning rate specification via local quadratic approximation." Neural Networks 141 (September 2021): 11–29. http://dx.doi.org/10.1016/j.neunet.2021.03.025.

Full text
APA, Harvard, Vancouver, ISO, and other styles
3

Leen, Todd K., Bernhard Schottky, and David Saad. "Optimal asymptotic learning rate: Macroscopic versus microscopic dynamics." Physical Review E 59, no. 1 (January 1, 1999): 985–91. http://dx.doi.org/10.1103/physreve.59.985.

Full text
APA, Harvard, Vancouver, ISO, and other styles
4

Kalvit, Anand, and Assaf Zeevi. "Dynamic Learning in Large Matching Markets." ACM SIGMETRICS Performance Evaluation Review 50, no. 2 (August 30, 2022): 18–20. http://dx.doi.org/10.1145/3561074.3561081.

Full text
Abstract:
We study a sequential matching problem faced by large centralized platforms where "jobs" must be matched to "workers" subject to uncertainty about worker skill proficiencies. Jobs arrive at discrete times (possibly in batches of stochastic size and composition) with "job-types" observable upon arrival. To capture the "choice overload" phenomenon, we posit an unlimited supply of workers where each worker is characterized by a vector of attributes (aka "worker-types") sampled from an underlying population-level distribution. The distribution as well as mean payoffs for possible workerjob type-pairs are unobservables and the platform's goal is to sequentially match incoming jobs to workers in a way that maximizes its cumulative payoffs over the planning horizon. We establish lower bounds on the regret of any matching algorithm in this setting and propose a novel rate-optimal learning algorithm that adapts to aforementioned primitives online. Our learning guarantees highlight a distinctive characteristic of the problem: achievable performance only has a second-order dependence on worker-type distributions; we believe this finding may be of interest more broadly.
APA, Harvard, Vancouver, ISO, and other styles
5

Zheng, Jiangbo, Yanhong Gan, Ying Liang, Qingqing Jiang, and Jiatai Chang. "Joint Strategy of Dynamic Ordering and Pricing for Competing Perishables with Q-Learning Algorithm." Wireless Communications and Mobile Computing 2021 (March 13, 2021): 1–19. http://dx.doi.org/10.1155/2021/6643195.

Full text
Abstract:
We use Machine Learning (ML) to study firms’ joint pricing and ordering decisions for perishables in a dynamic loop. The research assumption is as follows: at the beginning of each period, the retailer prices both the new and old products and determines how many new products to order, while at the end of each period, the retailer decides how much remaining inventory should be carried over to the next period. The objective is to determine a joint pricing, ordering, and disposal strategy to maximize the total expected discounted profit. We establish a decision model based on Markov processes and use the Q-learning algorithm to obtain a near-optimal policy. From numerical analysis, we find that (i) the optimal number of old products carried over to the next period depends on the upper quantitative bound for old inventory; (ii) the optimal prices for new products are positively related to potential demand but negatively related to the decay rate, while the optimal prices for old products have a positive relationship with both; and (iii) ordering decisions are unrelated to the quantity of old products. When the decay rate is low or the variable ordering cost is high, the optimal orders exhibit a trapezoidal decline as the quantity of new products increases.
APA, Harvard, Vancouver, ISO, and other styles
6

Chen, Zhigang, Rongwei Xu, and Yongxi Yi. "Dynamic Optimal Control of Transboundary Pollution Abatement under Learning-by-Doing Depreciation." Complexity 2020 (June 9, 2020): 1–17. http://dx.doi.org/10.1155/2020/3763684.

Full text
Abstract:
This paper analyzes a dynamic Stackelberg differential game model of watershed transboundary water pollution abatement and discusses the optimal decision-making problem under non-cooperative and cooperative differential game, in which the accumulation effect and depreciation effect of learning-by-doing pollution abatement investment are taken into account. We use dynamic optimization theory to solve the equilibrium solution of models. Through numerical simulation analysis, the path simulation and analysis of the optimal trajectory curves of each variable under finite-planning horizon and long-term steady state were carried out. Under the finite-planning horizon, the longer the planning period is, the lower the optimal emission rate is in equilibrium. The long-term steady-state game under cooperative decision can effectively reduce the amount of pollution emission. The investment intensity of pollution abatement in the implementation of non-cooperative game is higher than that of cooperative game. Under the long-term steady state, the pollution abatement investment trajectory of the cooperative game is relatively stable and there is no obvious crowding out effect. Investment continues to rise, and the optimal equilibrium level at steady state is higher than that under non-cooperative decision making. The level of decline in pollution stock under finite-planning horizon is not significant. Under the condition of long-term steady state, the trajectories of upstream and downstream pollution in the non-cooperative model and cooperative model are similar, but cooperative decision-making model is superior to the non-cooperative model in terms of the period of stabilization and steady state.
APA, Harvard, Vancouver, ISO, and other styles
7

De, Shipra, and Darryl A. Seale. "Dynamic Decision Making and Race Games." ISRN Operations Research 2013 (August 7, 2013): 1–15. http://dx.doi.org/10.1155/2013/452162.

Full text
Abstract:
Frequent criticism of dynamic decision making research pertains to the overly complex nature of the decision tasks used in experimentation. To address such concerns, we study dynamic decision making with respect to a simple race game, which has a computable optimal strategy. In this two-player race game, individuals compete to be the first to reach a designated threshold of points. Players alternate rolling a desired quantity of dice. If the number one appears on any of the dice, the player receives no points for his turn; otherwise, the sum of the numbers appearing on the dice is added to the player's score. Results indicate that although players are influenced by the game state when making their decisions, they tend to play too conservatively in comparison to the optimal policy and are influenced by the behavior of their opponents. Improvement in performance was negligible with repeated play. Survey data suggests that this outcome could be due to inadequate time for learning or insufficient player motivation. However, some players approached optimal heuristic strategies, which perform remarkably well.
APA, Harvard, Vancouver, ISO, and other styles
8

Yao, Yuhang, and Carlee Joe-Wong. "Interpretable Clustering on Dynamic Graphs with Recurrent Graph Neural Networks." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 5 (May 18, 2021): 4608–16. http://dx.doi.org/10.1609/aaai.v35i5.16590.

Full text
Abstract:
We study the problem of clustering nodes in a dynamic graph, where the connections between nodes and nodes' cluster memberships may change over time, e.g., due to community migration. We first propose a dynamic stochastic block model that captures these changes, and a simple decay-based clustering algorithm that clusters nodes based on weighted connections between them, where the weight decreases at a fixed rate over time. This decay rate can then be interpreted as signifying the importance of including historical connection information in the clustering. However, the optimal decay rate may differ for clusters with different rates of turnover. We characterize the optimal decay rate for each cluster and propose a clustering method that achieves almost exact recovery of the true clusters. We then demonstrate the efficacy of our clustering algorithm with optimized decay rates on simulated graph data. Recurrent neural networks (RNNs), a popular algorithm for sequence learning, use a similar decay-based method, and we use this insight to propose two new RNN-GCN (graph convolutional network) architectures for semi-supervised graph clustering. We finally demonstrate that the proposed architectures perform well on real data compared to state-of-the-art graph clustering algorithms.
APA, Harvard, Vancouver, ISO, and other styles
9

Liu, Haijun. "A Study of an IT-Assisted Higher Education Model Based on Distributed Hardware-Assisted Tracking Intervention." Occupational Therapy International 2022 (April 8, 2022): 1–12. http://dx.doi.org/10.1155/2022/8862716.

Full text
Abstract:
This paper presents an in-depth study and analysis of the model of higher education using distributed hardware tracking intervention of information technology. The MEC-based dynamic adaptive video stream caching technology model is proposed. The model dynamically adjusts the bit rate by referring to the broadband estimation and cache occupancy data to ensure users have a smooth experience effect. Simulation results show that the model has fewer transcoding times and generates lower latency than the traditional model, which is suitable for dual-teacher classroom scenarios and further improves the quality of the user’s video viewing experience. The model uses an edge cloud collaborative architecture to migrate the rendering technology to an edge server closer to the user side, enabling real-time interaction, computation, and rendering, reducing the time of data transmission as well as computation time. According to the blended learning-based adaptive intervention model, three rounds of teaching practice are conducted to validate the effectiveness of the intervention model in terms of both student process performance and outcome performance, thereby improving learning adaptability and improving learning effect. Teachers’ teaching has a significant impact on learning motivation ( β = 0.311 , p < 0.01 ), which in turn affects learning adaptability. Teachers use scientific teaching methods to stimulate students’ learning motivation, mobilize enthusiasm, and improve learning adaptability. Under the communication topology of the system as a directed graph, a multi-intelligent system dynamic model with grouping is established; i.e., the intragroup intelligence has the same dynamics but is different between groups, and all system dynamics are unknown. The proposed novel policy iterative algorithm is used to learn the optimal control protocol and achieve optimal consistency control. The effectiveness of the algorithm is demonstrated by the simulation experimental results. The simulation results show that the model has lower latency and energy consumption compared to the cloud rendering model, which is suitable for the safety education classroom scenario and solves the outstanding problems of network connection rate and cloud service latency.
APA, Harvard, Vancouver, ISO, and other styles
10

Li, Ao, Zhaoman Wan, and Zhong Wan. "Optimal Design of Online Sequential Buy-Price Auctions with Consumer Valuation Learning." Asia-Pacific Journal of Operational Research 37, no. 03 (April 29, 2020): 2050012. http://dx.doi.org/10.1142/s0217595920500128.

Full text
Abstract:
Buy-price auction has been successfully used as a new channel of online sales. This paper studies an online sequential buy-price auction problem, where a seller has an inventory of identical products and needs to clear them through a sequence of online buy-price auctions such that the total profit is maximized by optimizing the buy price in each auction. We propose a methodology by dynamic programming approach to solve this optimization problem. Since the consumers’ behavior affects the seller’s revenue, the consumers’ strategy used in this auction is first investigated. Then, two different dynamic programming models are developed to optimize the seller’s decision-making: one is the clairvoyant model corresponding to a situation where the seller has complete information about consumer valuations, and the other is the Bayesian learning model where the seller makes optimal decisions by continuously recording and utilizing auction data during the sales process. Numerical experiments are employed to demonstrate the impacts of several key factors on the optimal solutions, including the size of inventory, the number of potential consumers, and the rate at which the seller discounts early incomes. It is shown that when the consumers’ valuations are uniformly distributed, the Bayesian learning model is of great efficiency if the demand is adequate.
APA, Harvard, Vancouver, ISO, and other styles
More sources

Dissertations / Theses on the topic "Dynamic optimal learning rate"

1

Cheng, Martin Chun-Sheng, and pjcheng@ozemail com au. "Dynamical Near Optimal Training for Interval Type-2 Fuzzy Neural Network (T2FNN) with Genetic Algorithm." Griffith University. School of Microelectronic Engineering, 2003. http://www4.gu.edu.au:8080/adt-root/public/adt-QGU20030722.172812.

Full text
Abstract:
Type-2 fuzzy logic system (FLS) cascaded with neural network, called type-2 fuzzy neural network (T2FNN), is presented in this paper to handle uncertainty with dynamical optimal learning. A T2FNN consists of type-2 fuzzy linguistic process as the antecedent part and the two-layer interval neural network as the consequent part. A general T2FNN is computational intensive due to the complexity of type 2 to type 1 reduction. Therefore the interval T2FNN is adopted in this paper to simplify the computational process. The dynamical optimal training algorithm for the two-layer consequent part of interval T2FNN is first developed. The stable and optimal left and right learning rates for the interval neural network, in the sense of maximum error reduction, can be derived for each iteration in the training process (back propagation). It can also be shown both learning rates can not be both negative. Further, due to variation of the initial MF parameters, i.e. the spread level of uncertain means or deviations of interval Gaussian MFs, the performance of back propagation training process may be affected. To achieve better total performance, a genetic algorithm (GA) is designed to search better-fit spread rate for uncertain means and near optimal learnings for the antecedent part. Several examples are fully illustrated. Excellent results are obtained for the truck backing-up control and the identification of nonlinear system, which yield more improved performance than those using type-1 FNN.
APA, Harvard, Vancouver, ISO, and other styles
2

Cheng, Martin Chun-Sheng. "Dynamical Near Optimal Training for Interval Type-2 Fuzzy Neural Network (T2FNN) with Genetic Algorithm." Thesis, Griffith University, 2003. http://hdl.handle.net/10072/366350.

Full text
Abstract:
Type-2 fuzzy logic system (FLS) cascaded with neural network, called type-2 fuzzy neural network (T2FNN), is presented in this paper to handle uncertainty with dynamical optimal learning. A T2FNN consists of type-2 fuzzy linguistic process as the antecedent part and the two-layer interval neural network as the consequent part. A general T2FNN is computational intensive due to the complexity of type 2 to type 1 reduction. Therefore the interval T2FNN is adopted in this paper to simplify the computational process. The dynamical optimal training algorithm for the two-layer consequent part of interval T2FNN is first developed. The stable and optimal left and right learning rates for the interval neural network, in the sense of maximum error reduction, can be derived for each iteration in the training process (back propagation). It can also be shown both learning rates can not be both negative. Further, due to variation of the initial MF parameters, i.e. the spread level of uncertain means or deviations of interval Gaussian MFs, the performance of back propagation training process may be affected. To achieve better total performance, a genetic algorithm (GA) is designed to search better-fit spread rate for uncertain means and near optimal learnings for the antecedent part. Several examples are fully illustrated. Excellent results are obtained for the truck backing-up control and the identification of nonlinear system, which yield more improved performance than those using type-1 FNN.
Thesis (Masters)
Master of Philosophy (MPhil)
School of Microelectronic Engineering
Full Text
APA, Harvard, Vancouver, ISO, and other styles
3

Chang, Yusun. "Dynamic Optimal Fragmentation with Rate Adaptation in Wireless Mobile Networks." Diss., Georgia Institute of Technology, 2007. http://hdl.handle.net/1853/19824.

Full text
Abstract:
Dynamic optimal fragmentation with rate adaptation (DORA) is an algorithm to achieve maximum goodput in wireless mobile networks. With the analytical model that incorporates number of users, contentions, packet lengths, and bit error rates in the network, DORA computes a fragmentation threshold and transmits optimal sized packets with maximum rates. To estimate the SNR in the model, an adaptive on-demand UDP estimator is designed to reduce overheads. Test-beds to execute experiments for channel estimation, WLANs, Ad Hoc networks, and Vehicle-to-Vehicle networks are developed to evaluate the performance of DORA. DORA is an energy-efficient generic CSMA/CA MAC protocol for wireless mobile computing applications, and enhances system goodput in WLANs, Ad Hoc networks, and Vehicle-to-Vehicle networks without modification of the protocols.
APA, Harvard, Vancouver, ISO, and other styles
4

Moncur, Tyler. "Optimal Learning Rates for Neural Networks." BYU ScholarsArchive, 2020. https://scholarsarchive.byu.edu/etd/8662.

Full text
Abstract:
Neural networks have long been known as universal function approximators and have more recently been shown to be powerful and versatile in practice. But it can be extremely challenging to find the right set of parameters and hyperparameters. Model training is both expensive and difficult due to the large number of parameters and sensitivity to hyperparameters such as learning rate and architecture. Hyperparameter searches are notorious for requiring tremendous amounts of processing power and human resources. This thesis provides an analytic approach to estimating the optimal value of one of the key hyperparameters in neural networks, the learning rate. Where possible, the analysis is computed exactly, and where necessary, approximations and assumptions are used and justified. The result is a method that estimates the optimal learning rate for a certain type of network, a fully connected CReLU network.
APA, Harvard, Vancouver, ISO, and other styles
5

Shu, Weihuan. "Optimal sampling rate assignment with dynamic route selection for real-time wireless sensor networks." Thesis, McGill University, 2009. http://digitool.Library.McGill.CA:80/R/?func=dbin-jump-full&object_id=32351.

Full text
Abstract:
The allocation of computation and communication resources in a manner that optimizes aggregate system performance is a crucial aspect of system management. Wireless sensor network poses new challenges due to the resource constraints and real-time requirements. Existing work has dealt with the real-time sampling rate assignment problem, under single processor case and network case with static routing environment. For wireless sensor networks, in order to achieve better overall network performance, routing should be considered together with the rate assignments of individual flows. In this thesis, we address the problem of optimizing sampling rates with dynamic route selection for wireless sensor networks. We model the problem as a constrained optimization problem and solve it under the Network Utility Maximization framework. Based on the primal-dual method and dual decomposition technique, we design a distributed algorithm that achieves the optimal global network utility considering both dynamic route decision and rate assignment. Extensive simulations have been conducted to demonstrate the efficiency and efficacy of our proposed solutions.
L'attribution de calcul et de la communication ressources d'une mani`ere qui optimise les performances du syst`eme global est un aspect crucial de la gestion du syst`eme. R´eseau de capteurs sans fil pose de nouveaux d´efis en raison de la p´enurie de ressources et en temps r´eel. Travaux existants a traite distribution temps-reel probl`eme de taux d'´echantillonnage, dans un seul processeur cas et r´eseau cas de routage environment statique. Pour les r´eseaux de capteurs sans fil, afin de parvenir `a une meilleure performance globale du r´eseau, le routage devrait tre examin´e en mˆeme temps que la distribution de taux des flux individuels. Dans cet article, nous abordons le probl`eme de l'optimisation des taux d'´echantillonnage avec route s´election dynamique pour r´eseaux de capteurs sans fil. Nous modelisons le probleme comme un probl`eme d'optimisation et le r´esolvons dans le cadre de l'utilite de reseau maximisation. Sur la base de la m´ethode primal-dual et la dual d´ecomposition technique, nous concevons un algorithme distribu´e qui atteint le meilleur l'utilite de reseau globale au vu de route d´ecision dynamique et le taux distribution. Des simulations ont ´et´e r´ealis´ees pour d´emontrer l'efficience et l'efficacit´e de nos solutions propos´ees. fr
APA, Harvard, Vancouver, ISO, and other styles
6

Aroh, Kosisochukwu C. "Determination of optimal conditions and kinetic rate parameters in continuous flow systems with dynamic inputs." Thesis, Massachusetts Institute of Technology, 2018. https://hdl.handle.net/1721.1/121815.

Full text
Abstract:
Thesis: Ph. D., Massachusetts Institute of Technology, Department of Chemical Engineering, 2019
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 171-185).
.The fourth industrial revolution is said to be brought about by digitization in the manufacturing sector. According to this understanding, the third industrial revolution which involved computers and automation will be further enhanced with smart and autonomous systems fueled by data and machine learning. At the research stage, an analogous story is being told in how automation and new technologies could revolutionize a chemistry laboratory. Flow chemistry is a technique that contrast with traditional batch chemistry in one aspect as a method that facilitates process automation and in small scales, delivers process improvements such as high heat and mass transfer rates. In addition to flow chemistry, analytical tools have also greatly improved and have become fully automated with potential for remote control. Over the past decade, work utilizing optimization techniques to find optimal conditions in flow chemistry have become more prevalent.
In addition, the scope of reactions performed in these systems have also increased. In the first part of this thesis, the construction of a platform capable of performing a wide range of these reactions on the lab scale is discussed. This platform was built with the capability of performing global optimizations using steady state experiments. The rest of the thesis concerns generating dynamic experiments in flow systems and using these conditions to gain more information about a reaction. The ability to use dynamic experiments to accurately determine reaction kinetics is first detailed. Through these experiments we found that only two orthogonal experiments were needed to sample the experimental space. After this an algorithm that utilizes dynamic experiments for kinetic parameter estimation problems is described. The approach here was to use dynamic experiments to first quickly sample the design space to get a reasonable estimate of the kinetic parameters.
Then steady state optimal design of experiments were used to fine tune these estimates. We observed that after initial orthogonal experiments only three more conditions were needed for accurate estimates of the multi-step reaction example. In a similar fashion, an algorithm for reaction optimization that relies on dynamic experiments is also described. The approach here extended that of adaptive response surface methodology where dynamic orthogonal experiments were performed in place of steady state experiments. When compared to steady state optimizations of multi-step reactions, a reduction by half in time needed to locate the optimum is observed. Finally, the potential issues that arise when using transient experiments in automated systems for reaction analysis are addressed. These issues include dispersion, sampling rate, reactor sizes and the rate of change of transients.
These results demonstrate a way with which technological innovation could further revolutionize the chemistry laboratory. By combining machine learning, clouding computing and efficient, high information experiments reaction data could be quickly collected, and the information gained could be maximized for future predictions or optimizations.
by Kosisochukwu C. Aroh.
Ph. D.
Ph.D. Massachusetts Institute of Technology, Department of Chemical Engineering
APA, Harvard, Vancouver, ISO, and other styles
7

Ouyang, Hua. "Optimal stochastic and distributed algorithms for machine learning." Diss., Georgia Institute of Technology, 2013. http://hdl.handle.net/1853/49091.

Full text
Abstract:
Stochastic and data-distributed optimization algorithms have received lots of attention from the machine learning community due to the tremendous demand from the large-scale learning and the big-data related optimization. A lot of stochastic and deterministic learning algorithms are proposed recently under various application scenarios. Nevertheless, many of these algorithms are based on heuristics and their optimality in terms of the generalization error is not sufficiently justified. In this talk, I will explain the concept of an optimal learning algorithm, and show that given a time budget and proper hypothesis space, only those achieving the lower bounds of the estimation error and the optimization error are optimal. Guided by this concept, we investigated the stochastic minimization of nonsmooth convex loss functions, a central problem in machine learning. We proposed a novel algorithm named Accelerated Nonsmooth Stochastic Gradient Descent, which exploits the structure of common nonsmooth loss functions to achieve optimal convergence rates for a class of problems including SVMs. It is the first stochastic algorithm that can achieve the optimal O(1/t) rate for minimizing nonsmooth loss functions. The fast rates are confirmed by empirical comparisons with state-of-the-art algorithms including the averaged SGD. The Alternating Direction Method of Multipliers (ADMM) is another flexible method to explore function structures. In the second part we proposed stochastic ADMM that can be applied to a general class of convex and nonsmooth functions, beyond the smooth and separable least squares loss used in lasso. We also demonstrate the rates of convergence for our algorithm under various structural assumptions of the stochastic function: O(1/sqrt{t}) for convex functions and O(log t/t) for strongly convex functions. A novel application named Graph-Guided SVM is proposed to demonstrate the usefulness of our algorithm. We also extend the scalability of stochastic algorithms to nonlinear kernel machines, where the problem is formulated as a constrained dual quadratic optimization. The simplex constraint can be handled by the classic Frank-Wolfe method. The proposed stochastic Frank-Wolfe methods achieve comparable or even better accuracies than state-of-the-art batch and online kernel SVM solvers, and are significantly faster. The last part investigates the problem of data-distributed learning. We formulate it as a consensus-constrained optimization problem and solve it with ADMM. It turns out that the underlying communication topology is a key factor in achieving a balance between a fast learning rate and computation resource consumption. We analyze the linear convergence behavior of consensus ADMM so as to characterize the interplay between the communication topology and the penalty parameters used in ADMM. We observe that given optimal parameters, the complete bipartite and the master-slave graphs exhibit the fastest convergence, followed by bi-regular graphs.
APA, Harvard, Vancouver, ISO, and other styles
8

Lee, Jong Min. "A Study on Architecture, Algorithms, and Applications of Approximate Dynamic Programming Based Approach to Optimal Control." Diss., Georgia Institute of Technology, 2004. http://hdl.handle.net/1853/5048.

Full text
Abstract:
This thesis develops approximate dynamic programming (ADP) strategies suitable for process control problems aimed at overcoming the limitations of MPC, which are the potentially exorbitant on-line computational requirement and the inability to consider the future interplay between uncertainty and estimation in the optimal control calculation. The suggested approach solves the DP only for the state points visited by closed-loop simulations with judiciously chosen control policies. The approach helps us combat a well-known problem of the traditional DP called 'curse-of-dimensionality,' while it allows the user to derive an improved control policy from the initial ones. The critical issue of the suggested method is a proper choice and design of function approximator. A local averager with a penalty term is proposed to guarantee a stably learned control policy as well as acceptable on-line performance. The thesis also demonstrates versatility of the proposed ADP strategy with difficult process control problems. First, a stochastic adaptive control problem is presented. In this application an ADP-based control policy shows an "active" probing property to reduce uncertainties, leading to a better control performance. The second example is a dual-mode controller, which is a supervisory scheme that actively prevents the progression of abnormal situations under a local controller at their onset. Finally, two ADP strategies for controlling nonlinear processes based on input-output data are suggested. They are model-based and model-free approaches, and have the advantage of conveniently incorporating the knowledge of identification data distribution into the control calculation with performance improvement.
APA, Harvard, Vancouver, ISO, and other styles
9

Bountourelis, Theologos. "Efficient pac-learning for episodic tasks with acyclic state spaces and the optimal node visitation problem in acyclic stochastic digaphs." Diss., Atlanta, Ga. : Georgia Institute of Technology, 2008. http://hdl.handle.net/1853/28144.

Full text
Abstract:
Thesis (M. S.)--Industrial and Systems Engineering, Georgia Institute of Technology, 2009.
Committee Chair: Reveliotis, Spyros; Committee Member: Ayhan, Hayriye; Committee Member: Goldsman, Dave; Committee Member: Shamma, Jeff; Committee Member: Zwart, Bert.
APA, Harvard, Vancouver, ISO, and other styles
10

Singh, Manish Kumar. "Optimization, Learning, and Control for Energy Networks." Diss., Virginia Tech, 2021. http://hdl.handle.net/10919/104064.

Full text
Abstract:
Massive infrastructure networks such as electric power, natural gas, or water systems play a pivotal role in everyday human lives. Development and operation of these networks is extremely capital-intensive. Moreover, security and reliability of these networks is critical. This work identifies and addresses a diverse class of computationally challenging and time-critical problems pertaining to these networks. This dissertation extends the state of the art on three fronts. First, general proofs of uniqueness for network flow problems are presented, thus addressing open problems. Efficient network flow solvers based on energy function minimizations, convex relaxations, and mixed-integer programming are proposed with performance guarantees. Second, a novel approach is developed for sample-efficient training of deep neural networks (DNN) aimed at solving optimal network dispatch problems. The novel feature here is that the DNNs are trained to match not only the minimizers, but also their sensitivities with respect to the optimization problem parameters. Third, control mechanisms are designed that ensure resilient and stable network operation. These novel solutions are bolstered by mathematical guarantees and extensive simulations on benchmark power, water, and natural gas networks.
Doctor of Philosophy
Massive infrastructure networks play a pivotal role in everyday human lives. A minor service disruption occurring locally in electric power, natural gas, or water networks is considered a significant loss. Uncertain demands, equipment failures, regulatory stipulations, and most importantly complicated physical laws render managing these networks an arduous task. Oftentimes, the first principle mathematical models for these networks are well known. Nevertheless, the computations needed in real-time to make spontaneous decisions frequently surpass the available resources. Explicitly identifying such problems, this dissertation extends the state of the art on three fronts: First, efficient models enabling the operators to tractably solve some routinely encountered problems are developed using fundamental and diverse mathematical tools; Second, quickly trainable machine learning based solutions are developed that enable spontaneous decision making while learning offline from sophisticated mathematical programs; and Third, control mechanisms are designed that ensure a safe and autonomous network operation without human intervention. These novel solutions are bolstered by mathematical guarantees and extensive simulations on benchmark power, water, and natural gas networks.
APA, Harvard, Vancouver, ISO, and other styles
More sources

Books on the topic "Dynamic optimal learning rate"

1

Chater, Sean Christopher. The optimal rate of monetary growth in a dynamic macro model with a labour market distortion. [s.l.]: typescript, 1995.

Find full text
APA, Harvard, Vancouver, ISO, and other styles
2

Lin, Xiaofeng, Qinglai Wei, Ruizhuo Song, and Benkai Li. Self-Learning Optimal Control of Nonlinear Systems: Adaptive Dynamic Programming Approach. Springer, 2017.

Find full text
APA, Harvard, Vancouver, ISO, and other styles
3

Lin, Xiaofeng, Qinglai Wei, Ruizhuo Song, and Benkai Li. Self-Learning Optimal Control of Nonlinear Systems: Adaptive Dynamic Programming Approach. Springer, 2019.

Find full text
APA, Harvard, Vancouver, ISO, and other styles
4

Horneff, Vanya, Raimond Maurer, and Olivia S. Mitchell. How Persistent Low Expected Returns Alter Optimal Life Cycle Saving, Investment, and Retirement Behavior. Oxford University Press, 2018. http://dx.doi.org/10.1093/oso/9780198827443.003.0008.

Full text
Abstract:
This chapter explores how an environment of persistent low returns influences saving, investing, and retirement behaviors, compared to what in the past had been conceived of as ‘normal’ financial conditions. Using a calibrated life cycle dynamic model with realistic tax, minimum distribution, and social security benefit rules, we can mimic the large peak at the earliest claiming age at 62 that is seen in the data. Also in line with the evidence, our baseline results show a smaller second peak at the (system-defined) Full Retirement Age of 66. In the context of a zero-return environment, we show that workers will optimally devote more of their savings to non-retirement accounts and less to 401(k) accounts, since the relative appeal of investing in taxable versus tax-qualified retirement accounts is lower in a low return setting. Finally, we show that people claim social security benefits later in a low interest rate environment.
APA, Harvard, Vancouver, ISO, and other styles
5

Back, Kerry E. Real Options and q Theory. Oxford University Press, 2017. http://dx.doi.org/10.1093/acprof:oso/9780190241148.003.0020.

Full text
Abstract:
The theory of perpetual options and dynamic programming are applied to analyze the optimal capital investment of a firm. When investment is continuous and capital is the numeraire, the marginal value of capital is called marginal q. The optimal investment rate is a function of marginal q. When investment is irreversible and there is no depreciation, the optimal time to make each marginal investment is given by the theory of perpetual options. The optimal invesment times can also be calculated by dynamic programming. Fluctuations in marginal q add risk to a firm, compared to reversible investment. The Berk‐Green‐Naik model is an example of a model that relates risk and expected return to size and book‐to‐market by endogenizing investment.
APA, Harvard, Vancouver, ISO, and other styles
6

Hughes, Jim. Hip and femur. Oxford University Press, 2018. http://dx.doi.org/10.1093/med/9780198813170.003.0013.

Full text
Abstract:
Fractures of the hip are associated with various complications. Patients with hip fractures are often bed-bound and require intensive nursing and care. The rate of hip fractures is likely to continue to rise globally, and early intervention is currently seen as the optimal course for patients with such injuries. This chapter covers a selection of orthopaedic procedures involving the hip and femur, covering cannulated hip screws, dynamic hip screws, antegrade and retrograde femoral intra-medullary nailing, and plating of the midshaft of the femur. Each procedure includes images that demonstrate the position of the C-arm, patient, and surgical equipment, with accompanying radiographs demonstrating the resulting images.
APA, Harvard, Vancouver, ISO, and other styles
7

Björk, Tomas. Arbitrage Theory in Continuous Time. Oxford University Press, 2019. http://dx.doi.org/10.1093/oso/9780198851615.001.0001.

Full text
Abstract:
The fourth edition of this textbook on pricing and hedging of financial derivatives, now also including dynamic equilibrium theory, continues to combine sound mathematical principles with economic applications. Concentrating on the probabilistic theory of continuous time arbitrage pricing of financial derivatives, including stochastic optimal control theory and optimal stopping theory, the book is designed for graduate students in economics and mathematics, and combines the necessary mathematical background with a solid economic focus. It includes a solved example for every new technique presented, contains numerous exercises, and suggests further reading in each chapter. All concepts and ideas are discussed, not only from a mathematics point of view, but the mathematical theory is also always supplemented with lots of intuitive economic arguments. In the substantially extended fourth edition Tomas Björk has added completely new chapters on incomplete markets, treating such topics as the Esscher transform, the minimal martingale measure, f-divergences, optimal investment theory for incomplete markets, and good deal bounds. There is also an entirely new part of the book presenting dynamic equilibrium theory. This includes several chapters on unit net supply endowments models, and the Cox–Ingersoll–Ross equilibrium factor model (including the CIR equilibrium interest rate model). Providing two full treatments of arbitrage theory—the classical delta hedging approach and the modern martingale approach—the book is written in such a way that these approaches can be studied independently of each other, thus providing the less mathematically oriented reader with a self-contained introduction to arbitrage theory and equilibrium theory, while at the same time allowing the more advanced student to see the full theory in action.
APA, Harvard, Vancouver, ISO, and other styles
8

Vervaeke, John, Leo Ferraro, and Arianne Herrera-Bennett. Flow as Spontaneous Thought. Edited by Kalina Christoff and Kieran C. R. Fox. Oxford University Press, 2018. http://dx.doi.org/10.1093/oxfordhb/9780190464745.013.8.

Full text
Abstract:
Flow is an experience encountered in many areas of human endeavor; it is reported by athletes and artists, writers and thinkers. Paradoxically, it appears to involve significant energy expenditure, and yet it is reported to feel almost effortless. It is a prototypical instance of spontaneous thought. The flow experience has been extensively documented and studied by many scholars, most prominently Csikszentmihalyi, who characterized it as “optimal experience.” This chapter builds on the work of Csikszentmihalyi and others by providing a cognitive scientific account of flow, a framework that organizes and integrates the various cognitive processes and features that serve to make flow an optimal experience. In particular, it is argued that flow is characterized by a dynamic cascade of insight, coupled with enhanced implicit learning. This model seeks to integrate the phenomenological accounts of flow with the existing body of cognitive research.
APA, Harvard, Vancouver, ISO, and other styles

Book chapters on the topic "Dynamic optimal learning rate"

1

Powell, Warren B., and Ilya O. Ryzhov. "Optimal Learning and Approximate Dynamic Programming." In Reinforcement Learning and Approximate Dynamic Programming for Feedback Control, 410–31. Hoboken, NJ, USA: John Wiley & Sons, Inc., 2013. http://dx.doi.org/10.1002/9781118453988.ch18.

Full text
APA, Harvard, Vancouver, ISO, and other styles
2

Kim, Minjung, and Yucel Altunbasak. "Optimal Dynamic Rate Shaping for Compressed Video Streaming." In Networking — ICN 2001, 786–94. Berlin, Heidelberg: Springer Berlin Heidelberg, 2001. http://dx.doi.org/10.1007/3-540-47734-9_78.

Full text
APA, Harvard, Vancouver, ISO, and other styles
3

Wei, Qinglai, Ruizhuo Song, Benkai Li, and Xiaofeng Lin. "Principle of Adaptive Dynamic Programming." In Self-Learning Optimal Control of Nonlinear Systems, 1–17. Singapore: Springer Singapore, 2017. http://dx.doi.org/10.1007/978-981-10-4080-1_1.

Full text
APA, Harvard, Vancouver, ISO, and other styles
4

Dvijotham, K., and E. Todorov. "Linearly Solvable Optimal Control." In Reinforcement Learning and Approximate Dynamic Programming for Feedback Control, 119–41. Hoboken, NJ, USA: John Wiley & Sons, Inc., 2013. http://dx.doi.org/10.1002/9781118453988.ch6.

Full text
APA, Harvard, Vancouver, ISO, and other styles
5

You, Shuai, Wanyi Gao, Ziyang Li, Qifen Yang, Meng Tian, and Shuhua Zhu. "Dynamic Adjustment of the Learning Rate Using Gradient." In Human Centered Computing, 61–69. Cham: Springer Nature Switzerland, 2022. http://dx.doi.org/10.1007/978-3-031-23741-6_6.

Full text
APA, Harvard, Vancouver, ISO, and other styles
6

Kuflik, Tsvi, Bracha Shapira, Yuval Elovici, and Adlai Maschiach. "Privacy Preservation Improvement by Learning Optimal Profile Generation Rate." In User Modeling 2003, 168–77. Berlin, Heidelberg: Springer Berlin Heidelberg, 2003. http://dx.doi.org/10.1007/3-540-44963-9_23.

Full text
APA, Harvard, Vancouver, ISO, and other styles
7

Fairbank, Michael, Danil Prokhorov, and Eduardo Alonso. "Approximating Optimal Control with Value Gradient Learning." In Reinforcement Learning and Approximate Dynamic Programming for Feedback Control, 142–61. Hoboken, NJ, USA: John Wiley & Sons, Inc., 2013. http://dx.doi.org/10.1002/9781118453988.ch7.

Full text
APA, Harvard, Vancouver, ISO, and other styles
8

Vamvoudakis, Kyriakos G., and Frank L. Lewis. "Online Learning Algorithms for Optimal Control and Dynamic Games." In Reinforcement Learning and Approximate Dynamic Programming for Feedback Control, 350–77. Hoboken, NJ, USA: John Wiley & Sons, Inc., 2013. http://dx.doi.org/10.1002/9781118453988.ch16.

Full text
APA, Harvard, Vancouver, ISO, and other styles
9

Liu, Derong, Qinglai Wei, Ding Wang, Xiong Yang, and Hongliang Li. "Learning Algorithms for Differential Games of Continuous-Time Systems." In Adaptive Dynamic Programming with Applications in Optimal Control, 417–80. Cham: Springer International Publishing, 2017. http://dx.doi.org/10.1007/978-3-319-50815-3_11.

Full text
APA, Harvard, Vancouver, ISO, and other styles
10

Rafajłowicz, Ewaryst, and Wojciech Rafajłowicz. "Iterative Learning in Repetitive Optimal Control of Linear Dynamic Processes." In Artificial Intelligence and Soft Computing, 705–17. Cham: Springer International Publishing, 2016. http://dx.doi.org/10.1007/978-3-319-39378-0_60.

Full text
APA, Harvard, Vancouver, ISO, and other styles

Conference papers on the topic "Dynamic optimal learning rate"

1

Roy, Serge. "Near-optimal dynamic learning rate for training backpropagation neural networks." In Optical Engineering and Photonics in Aerospace Sensing, edited by Dennis W. Ruck. SPIE, 1993. http://dx.doi.org/10.1117/12.152627.

Full text
APA, Harvard, Vancouver, ISO, and other styles
2

Zhang, Tong, C. L. Philip Chen, Chi-Hsu Wang, and Sik Chung Tam. "A new dynamic optimal learning rate for a two-layer neural network." In 2012 International Conference on System Science and Engineering (ICSSE). IEEE, 2012. http://dx.doi.org/10.1109/icsse.2012.6257148.

Full text
APA, Harvard, Vancouver, ISO, and other styles
3

Tong, Zhang, C. L. Philip Chen, and Zhou Jin. "Impact of ratio k on two-layer neural networks with dynamic optimal learning rate." In 2014 International Joint Conference on Neural Networks (IJCNN). IEEE, 2014. http://dx.doi.org/10.1109/ijcnn.2014.6889774.

Full text
APA, Harvard, Vancouver, ISO, and other styles
4

Viaña Perez, Javier, Drew Scott, Manish Kumar, and Kelly Cohen. "Dynamic Genetic Algorithm for Optimizing Movement of a Six-Limb Creature." In ASME 2020 Dynamic Systems and Control Conference. American Society of Mechanical Engineers, 2020. http://dx.doi.org/10.1115/dscc2020-3243.

Full text
Abstract:
Abstract In this study we consider a Dynamic Genetic Algorithm used to optimize the movement of a symmetric six-legged creature. The optimal movement is that which advances the creature in a straight line forward with the greatest average speed. The mutation rate and crossover rate are adjusted based on number of iterations the algorithm has completed. This dynamic element was added to improve convergence rate as well as reducing the chance that the algorithm is stuck in a local optimum. The chromosomes are represented by a 2-dimensional array, where the rows represent sequences of movement. Each row defines the change in the angle for all the joints. Angular rates are restricted per joint, as well as ranges of motion. The fitness of a chromosome is determined by the resultant average speed, calculated as total displacement of the center of gravity over total time of movements in the chromosome. The results of this study show the possibility to breed mathematically the creature by using the Dynamic Genetic Algorithm proposed. This learning process converged, for all the simulations carried out, to the natural motion of six-legged beings like the ants.
APA, Harvard, Vancouver, ISO, and other styles
5

Cao, Jie, Hui Ren, and Dan Sui. "Global Optimization Workflow for Offshore Drilling Rate of Penetration With Dynamic Drilling Log Data." In ASME 2022 41st International Conference on Ocean, Offshore and Arctic Engineering. American Society of Mechanical Engineers, 2022. http://dx.doi.org/10.1115/omae2022-79747.

Full text
Abstract:
Abstract The prediction and optimization of drilling rate of penetration (ROP) are among the most effective approaches in improving drilling efficiency. To achieve that, it calls for a well-established prediction model and a well-defined optimization methodology. With the advancement in large dataset acquisition and computational efficient machine learning methods, data-driven ROP prediction has superiority over classical physical models. Furthermore, when the ROP prediction model is trained and validated, it can be used to optimize the controllable parameters, preferably globally, given objective functions and proper constraints. The global optimization of drilling ROP is desirable in the design phase, such that the controllable parameters can be optimized for the whole planned well depth. This provides an optimum plan pushing the limit of drilling efficiency and provides valuable controlling strategies that guide the drilling operations. The main object of this research is to investigate the global optimization workflow for ROP using prediction models based on machine learning methods. We first present an automated data processing method, dealing with and taking advantage of the variety and a vast amount of the drilling dataset. Then, the deep neural network (DNN) model for ROP predictions is validated and tested. In the trained predictive model, there are three controllable parameters, weight on bit (WOB), drilling string revolution speed (RPM), and drilling fluid flow rates (Q). Next, we choose the genetic algorithm (GA) to search the global optimal parameter combination in the control parameters space. The optimization workflow can be applied for the whole well depth, various segments of depth intervals, and different formation layers, resulting in a combination of controllable parameters for the entire well, for every section of given depth intervals, and for each formation layer, respectively. In summary, the global optimization workflow incorporates end-to-end data processing and promotes improved drilling efficiency. The global optimized results push the limit of drilling efficiency and provide valuable post-drilling analysis and offset drilling design recommendations. However, the extreme optimum results may not be reached in field practice, as more constraints such as formation information need to be applied to make the operation realistic.
APA, Harvard, Vancouver, ISO, and other styles
6

Trogmann, Hannes, Harald Waschl, Daniel Alberer, and Bernhard Spiegl. "Time Optimal Compressor Valve Soft Landing by Two Step ILC." In ASME 2011 Dynamic Systems and Control Conference and Bath/ASME Symposium on Fluid Power and Motion Control. ASMEDC, 2011. http://dx.doi.org/10.1115/dscc2011-6199.

Full text
Abstract:
Normally compressor load control is realized either passively with a bypass loop with high energy consumption or by an active valve. Actively controlled valves must fulfill a set of conditions which can be expressed as a constrained optimal control. However, compressor valves are subject to serious disturbances which might be unpredictable and difficult to describe, and have complex dynamics. This makes the solution of the real optimal problem difficult and even questionable. Against this background, this paper proposes a two step design approach: first a problem approximation is derived for which an explicit solution of the problem can be computed, then iterative learning control (ILC) is used with the real plant to enforce it. Due to limitation of the actuators, basically a switching solution results, and the tracking of this solution yields a satisfactory solution near to the optimal one. The method has been tested in simulation and experimentally on a special valve design including a small rotational electrical motor and a high gear rate spindle.
APA, Harvard, Vancouver, ISO, and other styles
7

Amadi, Kingsley Williams, Ibiye Iyalla, Prabhua Radhakrishna, Mortadha Torki Al Saba, and Marwa Mustapha Waly. "Continuous Dynamic Drill-Off Test Whilst Drilling Using Reinforcement Learning in Autonomous Rotary Drilling System." In ADIPEC. SPE, 2022. http://dx.doi.org/10.2118/211723-ms.

Full text
Abstract:
Abstract In the development of autonomous downhole drilling systems, decision-making in the selection of optimized operating parameters has become one of the technical difficulties. Traditionally, the driller performs a trial-and-error approach in search of optimal parameters, which is now less effective and non-sustainable to the changing drilling environment. This paper presents a decision-making process using reinforcement Q-learning algorithms that can act as real-time optimization algorithms for selecting an optimal operating parameter for rotary drilling systems using Q-learning on experimental published data from the literature. The reinforcement learning framework is a stochastic approximate dynamic programming, with varying estimation techniques for goal-directed sequential learning from interaction paradigms. First, a Markov Decision Process (MDP) is established by analyzing agent exploration and exploitation of possible actions taken in an environment. Second, the state set and action set are designed by the synthesized consideration of surface operating parameters from the published data within the range of operational limit. Then, sequentially, at each timestep, the agent takes an action (e.g., changing rotary speed or changing axial force) that makes the environment (formation) transition from one state to another. Consequently, the agent receives a reward (e.g., distance drilled) before taking the next action. Furthermore, a recursive reinforcement Q-learning algorithm is developed mainly based on the reward function and update function. Analysis of experimental data on drilling was implemented for five states of axial force parameters with five feed rate decisions on each of the states, whilst having distance of a hole drilled as a reward. The proposed optimization model computed using value iteration showed that following Decision 2 yielded the best result. The analysis results also revealed that the optimal value function was reached irrespective of the initial state conditions. The agent's objective is to learn policy mapping from states to actions such that the agent's cumulative reward (footage drilled) is maximized. The result of this research could be used as a decision-making tool in drilling operations that provides an engineered approach for optimal operating parameter selection and improvement in the efficiency of the drilling process in terms of cost and time.
APA, Harvard, Vancouver, ISO, and other styles
8

Narayan, Meenakshi, and Ann Majewicz Fey. "A Novel Approach to Time Series Forecasting Using Model-Free Adaptive Control Framework." In ASME 2020 Dynamic Systems and Control Conference. American Society of Mechanical Engineers, 2020. http://dx.doi.org/10.1115/dscc2020-3329.

Full text
Abstract:
Abstract Sensor data predictions could significantly improve the accuracy and effectiveness of modern control systems; however, existing machine learning and advanced statistical techniques to forecast time series data require significant computational resources which is not ideal for real-time applications. In this paper, we propose a novel forecasting technique called Compact Form Dynamic Linearization Model-Free Prediction (CFDL-MFP) which is derived from the existing model-free adaptive control framework. This approach enables near real-time forecasts of seconds-worth of time-series data due to its basis as an optimal control problem. The performance of the CFDL-MFP algorithm was evaluated using four real datasets including: force sensor readings from surgical needle, ECG measurements for heart rate, and atmospheric temperature and Nile water level recordings. On average, the forecast accuracy of CFDL-MFP was 28% better than the benchmark Autoregressive Integrated Moving Average (ARIMA) algorithm. The maximum computation time of CFDL-MFP was 49.1ms which was 170 times faster than ARIMA. Forecasts were best for deterministic data patterns, such as the ECG data, with a minimum average root mean squared error of (0.2±0.2).
APA, Harvard, Vancouver, ISO, and other styles
9

Docherty, David, Dale Erickson, and Scott Henderson. "Using Ai to Optimize the Use of Gas Lift in Oil Wells." In ADIPEC. SPE, 2022. http://dx.doi.org/10.2118/211028-ms.

Full text
Abstract:
Abstract Optimization of gas lift rates in an oil field with hundreds of wells is a complex challenge. Typically traditional control systems and operating strategies fail to optimize the problem due to this complexity. Typical gas lift optimization challenges in a field include the following: Real-time flow rate prediction of various phasesReal-time wells performanceReal-time pipeline network performance The dynamic nature of the problem and the variability of the solution space makes it extremely hard for traditional simulation-based solutions to locate the optimal performance point in real-time. The computational requirement is massive and makes it difficult to perform these calculations at the "edge". This is where combining simulations, human expertise, and machine learning technologies such as Deep Reinforcement Learning help build AI that can excel in rapidly computing optimized setpoints in complex domains. Using pioneering machine teaching methods combined with multiphase simulations this paper will present the solution of using Artificial Intelligence (AI) to optimize gas lift rates in real-time, finding the optimum gas lift rates for a 4 well pad, 200 well system, such that net profit is increased by 5%-25% for different baselines as the reservoir conditions change over field life.
APA, Harvard, Vancouver, ISO, and other styles
10

Fang, Haowen, Amar Shrestha, Ziyi Zhao, and Qinru Qiu. "Exploiting Neuron and Synapse Filter Dynamics in Spatial Temporal Learning of Deep Spiking Neural Network." In Twenty-Ninth International Joint Conference on Artificial Intelligence and Seventeenth Pacific Rim International Conference on Artificial Intelligence {IJCAI-PRICAI-20}. California: International Joint Conferences on Artificial Intelligence Organization, 2020. http://dx.doi.org/10.24963/ijcai.2020/388.

Full text
Abstract:
The recently discovered spatial-temporal information processing capability of bio-inspired Spiking neural networks (SNN) has enabled some interesting models and applications. However designing large-scale and high-performance model is yet a challenge due to the lack of robust training algorithms. A bio-plausible SNN model with spatial-temporal property is a complex dynamic system. Synapses and neurons behave as filters capable of preserving temporal information. As such neuron dynamics and filter effects are ignored in existing training algorithms, the SNN downgrades into a memoryless system and loses the ability of temporal signal processing. Furthermore, spike timing plays an important role in information representation, but conventional rate-based spike coding models only consider spike trains statistically, and discard information carried by its temporal structures. To address the above issues, and exploit the temporal dynamics of SNNs, we formulate SNN as a network of infinite impulse response (IIR) filters with neuron nonlinearity. We proposed a training algorithm that is capable to learn spatial-temporal patterns by searching for the optimal synapse filter kernels and weights. The proposed model and training algorithm are applied to construct associative memories and classifiers for synthetic and public datasets including MNIST, NMNIST, DVS 128 etc. Their accuracy outperforms state-of-the-art approaches.
APA, Harvard, Vancouver, ISO, and other styles

Reports on the topic "Dynamic optimal learning rate"

1

Araya, Mesele, Caine Rolleston, Pauline Rose, Ricardo Sabates, Dawit Tibebu Tiruneh, and Tassew Woldehanna. Understanding the Impact of Large-Scale Educational Reform on Students’ Learning Outcomes in Ethiopia: The GEQIP-II Case. Research on Improving Systems of Education (RISE), January 2023. http://dx.doi.org/10.35489/bsg-rise-wp_2023/125.

Full text
Abstract:
The Ethiopian education system has been very dynamic over recent years, with a series of large-scale education program interventions, such as the Second Phase of General Education Quality Improvement Project (GEQIP-II) that aimed to improve student learning outcomes. Despite the large-scale programs, empirical studies assessing how such interventions have worked and who benefited from the reforms are limited. This study aims to understand the impact of the reform on Grade 4 students’ maths learning outcomes over a school year using two comparable Grade 4 cohort students from 33 common schools in the Young Lives (YL, 2012-13) and RISE (2018-19) surveys. We employ matching techniques to estimate the effects of the reform by accounting for baseline observable characteristics of the two cohorts matched within the same schools. Results show that the RISE cohort started the school year with a lower average test score than the YL cohort. At the start of Grade 4, the Average Treatment Effect on the Treated (ATT) is lower by 0.36 SD (p<0.01). In terms of learning gain over the school year, however, the RISE cohort has shown a modestly higher value-added than the YL cohort, with ATT of 0.074 SD (p<0.05). The learning gain particularly is higher for students in rural schools (0.125 SD & p<0.05), which is also stronger among rural boys (0.184 SD & p<0.05) than among rural girls. We consider the implications of our results from a system dynamic perspective; in that the GEQIP-II reform induced unprecedented access to primary education, where the national Net Enrolment Rate (NER) rose from 85.7 percent in 2012-13 to 95.3 percent in 2019-20, which is equivalent to nearly 3 million additional learners to the primary education at a national level. This shows that learning levels have not increased in tandem with enrolment, and the unprecedented access for nearly all children might create pressure on the school system. Current policy efforts should therefore focus on sustaining learning gains for all children while creating better access.
APA, Harvard, Vancouver, ISO, and other styles
You might also be interested in the extended bibliographies on the topic 'Dynamic optimal learning rate' for particular source types:
Journal articles Dissertations / Theses
We offer discounts on all premium plans for authors whose works are included in thematic literature selections. Contact us to get a unique promo code!

To the bibliography