با سلام خدمت کاربران در صورتی که با خطای سیستم پرداخت بانکی مواجه شدید از طریق کارت به کارت (6037997535328901 بانک ملی ناصر خنجری ) مقاله خود را دریافت کنید (تا مشکل رفع گردد).
ردیف | عنوان | نوع |
---|---|---|
1 |
Integral reinforcement learning-based online adaptive event-triggered control for non-zero-sum games of partially unknown nonlinear systems
یادگیری تقویتی یکپارچه مبتنی بر رویداد انطباقی انلاین برای بازی های غیر مجموع صفر از سیستم های غیر خطی ناشناخته جزیی-2020 This paper develops an integral reinforcement learning (IRL)-based adaptive control method for the multi- player non-zero-sum (NZS) games of the nonlinear continuous-time systems with partially unknown dy- namics, in the context of event-triggered mechanism. With the principle of IRL method, the requirement for the system drift dynamics is relaxed in the controller design. Moreover, different from the conven- tional iteration computation methods, the algorithm developed in this work is implemented in an online adaptive fashion, which provides a new way to combine the IRL algorithm and the event-triggered con- trol framework in solving the NZS game issues. In the event-based algorithm, a state-dependent trigger- ing condition is presented, which not only guarantees the closed-loop system stability, but also reduces the computation and communication loads of the controlled plant. By means of Lyapunov theorem, the uniform ultimate boundedness (UUB) properties of the system states and the critic weight estimation errors have been proved. Finally, two numerical examples are utilized to demonstrate the efficacy of the proposed method. Keywords: Event-triggered control (ETC) | Integral reinforcement learning (IRL) | Adaptive dynamic programming (ADP) | Adaptive critic design | Non-zero-sum (NZS) games |
مقاله انگلیسی |
2 |
MRCDRL: Multi-robot coordination with deep reinforcement learning
MRCDRL: هماهنگی چند ریات با یادگیری تقویتی عمیق-2020 This paper proposes a multi-robot cooperative algorithm based on deep reinforcement learning (MR- CDRL). We use end-to-end methods to train directly from each robot-centered, relative perspective- generated image, and each robot’s reward as the input. During training, it is not necessary to specify the target position and movement path of each robot. MRCDRL learns the actions of each robot by training the neural network. MRCDRL uses the neural network structure that was modified from the Duel neural network structure. In the Duel network structure, there are two streams that each represents the state value function and the state-dependent action advantage function, and the results of the two streams are merged. The proposed method can solve the resource competition problem on the one hand and can solve the static and dynamic obstacle avoidance problems between multi-robot in real time on the other hand. Our new MRCDRL algorithm has higher accuracy and robustness than DQN and DDQN and can be effectively applied to multi-robot collaboration. Keywords: Cooperative control | Machine learning | Image processing |
مقاله انگلیسی |
3 |
Multistability of switched neural networks with sigmoidal activation functions under state-dependent switching
قابلیت چند منظوره تغییر شبکه های عصبی با توابع فعال سازی سیگموئید تحت تعویض وابسته به حالت-2020 This paper presents theoretical results on the multistability of switched neural networks with commonly
used sigmoidal activation functions under state-dependent switching. The multistability analysis
with such an activation function is difficult because state–space partition is not as straightforward as
that with piecewise-linear activations. Sufficient conditions are derived for ascertaining the existence
and stability of multiple equilibria. It is shown that the number of stable equilibria of an n-neuron
switched neural networks is up to 3n under given conditions. In contrast to existing multistability
results with piecewise-linear activation functions, the results herein are also applicable to the equilibria
at switching points. Four examples are discussed to substantiate the theoretical results. Keywords: Multistability | Switched neural network | State-dependent | Sigmoidal activation function |
مقاله انگلیسی |
4 |
Robust stability analysis of stochastic switched neural networks with parameter uncertainties via state-dependent switching law
تجزیه و تحلیل پایداری قوی شبکه های عصبی روشن تصادفی با عدم قطعیت پارامتر از طریق تغییر قانون وابسته به دولت-2020 The problem of robust stability analysis for a class of stochastic switched neural networks (SSNNs) with
time-varying parametric uncertainties is investigated in this paper. Some sufficient conditions are
derived to guarantee the robust global asymptotical stability in mean square for the uncertain SSNNs
by using state-dependent switching (SDS) method. It is shown that the robust stability of uncertain
SSNNs composed of all unstable subnetworks can be achieved by using the designed SDS law.
Moreover, the proposed sufficient conditions can be easily checked in terms of linear matrix inequalities
(LMIs) for conveniently using Matlab toolbox. A numerical example is provided to demonstrate the effectiveness
of the proposed SDS law. Keywords: Stochastic switched neural networks | Uncertainties | State-dependent switching | Linear matrix inequality | Globally asymptotical stability |
مقاله انگلیسی |
5 |
Origin and dynamics of cortical slow oscillations
منشا و پویایی نوسانات آهسته قشر مغز-2020 Slow oscillations are the coordinated activity of large neuronal
populations consisting of alternating active (Up states) and
silent periods (Down states). These oscillations occur in the
corticothalamocortical network during slow-wave sleep and
deep anesthesia. They also spontaneously occur in isolated
cortical slices or in disconnected ‘cortical islands’ in brain
damage. This rhythmic activity emerges in the cortical network
when there are no other driving inputs and is considered its
default activity pattern. During Up states, neocortical neurons
receive barrages of synaptic inputs and fire action potentials.
During Down states, neurons remain silent; rather they are
hyperpolarized, and synaptic activity is almost nonexistent.
From a dynamic perspective, this pattern is often referred to as
a state-dependent bistability. During Up states, the activity
expresses coherent oscillations at high frequencies in the beta
and gamma ranges, sharing properties with wakefulness. The
impact of Up/Down states on synaptic transmission and
plasticity and its relationship with sleep are discussed. |
مقاله انگلیسی |
6 |
Are gold and silver cointegrated? New evidence from quantile cointegrating regressions
آیا طلا و نقره هم - یکپارچه هستند؟ شواهد جدیدی از رگراسیون های هم - یکپارچگی چارک-2018 This paper revisits the study on the long-run relationship between gold and silver by Escribano and Granger (1998). We apply a quantile cointegration model to gold and silver prices and to prices of the corresponding futures contracts. Whereas cointegration models, assuming a constant cointegrating vector, fail to detect a cointegration relationship between gold and silver, we are able to show that a nonlinear long-run relationship exits. The cointegrating vector is modelled as state-dependent and time-varying in our framework and the quantile cointegration estimates reveal substantial asymmetry in the relationship. The results suggest that the pronounced role of precious metals as investment opportunities particularly in bubble-like episodes and times of financial turmoil leads to comovement of gold and silver in these periods.
keywords: Gold |Silver |Quantile cointegration |Time-varying |State-dependence |
مقاله انگلیسی |
7 |
سیستم های تشویقی برای تصمیمات سرمایه گذاری مخاطره آمیز تحت نظر تنظیمات ناشناخته
سال انتشار: 2017 - تعداد صفحات فایل pdf انگلیسی: 8 - تعداد صفحات فایل doc فارسی: 28 این مقاله به بررسی چگونگی طراحی سیستم هایی برای ایجاد انگیزه در مدیران برای تصمیم گیری سرمایه گذاری خطرناک چندین ساله می پردازد. ما نشان می دهیم که جبران کارکرد و کارایی عملکردی باید طراحی گردد تا مدیران در مجموعه پروژه ها برای به حداکثر رساندن ارزش مورد انتظار را اجرا اطمینان حاصل کنند. طرح ۱ (RBCA) و گسترش آن در مقالات مربوط به تنظیمات زمانی ناشناخته، به طور کلی در زمان های ناشناخته و ریسک پذیری به کار خود ادامه نمی دهند. ما نشان می دهیم که در مواجهه با چنین تنظیم های نامعلومی در یک محیط خطرناک، یک قانون تخصیص وابسته به حالت خاص مورد نیاز است. ما چنین طرح تخصیصی را معرفی می کنیم، که از ان به عنوان طرح RBCA مشروح دولت یاد می کنیم و نشان می دهیم؛ که دانش خاصی از زمان و ساختار ریسک جریان های نقدی برای اعمال آن لازم است.
کلمات کلیدی: حسابداری تعهدی | سازگاری | تخصیص هزینه | سیستم تشویقی | اندازه گیری عملکرد | RBCA |
مقاله ترجمه شده |