In the present paper a newer application of Artificial Neural Network (ANN) has been developed i.e., predicting response-function results of electrical-mechanical system through ANN. This method is specially useful to complex systems for which it is not possible to find the response-function because of complexity of the system. The proposed approach suggests that how even without knowing the response-function, the response-function results can be predicted with the use of ANN to the system. The steps used are: (i) Depending on the system, the ANN-architecture and the input & output parameters are decided, (ii) Training & test data are generated from simplified circuits and through tactic-superposition of it for complex circuits, (iii) Training the ANN with training data through many cycles and (iv) Test-data are used for predicting the response-function results. It is found that the proposed novel method for response prediction works satisfactorily. Thus this method could be used specially for complex systems where other methods are unable to tackle it. In this paper the application of ANN is particularly demonstrated to electrical-circuit system but can be applied to other systems too.

R. C. Gupta Ankur Agarwal Ruchi Gupta Sanjay Gupta

We give some additions to the article "On the generalized Riemann-Hilbert problem with irregular singularities" by Bolibruch, Malek, Mitschi (math/0410483). In particular, a weak GRH-problem and the GRH-problem for scalar differential equations are discussed.

R. R. Gontsov I. V. Vyugin

Artificial neural network (ANN) potentials enable highly accurate atomistic simulations of complex materials at unprecedented scales. Despite their promise, training ANN potentials to represent intricate potential energy surfaces (PES) with transferability to diverse chemical environments remains computationally intensive, especially when atomic force data are incorporated to improve PES gradients. Here, we present an efficient ANN potential training methodology that uses Gaussian process regression (GPR) to incorporate atomic forces into ANN training, leading to accurate PES models with fewer additional first-principles calculations and a reduced computational effort for training. Our GPR-ANN approach generates synthetic energy data from force information in the reference dataset, thus augmenting the training datasets and bypassing direct force training. Benchmark tests on hybrid density-functional theory data for ethylene carbonate (EC) molecules and Li metal-EC interfaces, relevant for lithium metal battery applications, demonstrate that GPR-ANN potentials achieve accuracies comparable to fully force-trained ANNs with a significantly reduced computational overhead. Detailed comparisons show that the method improves both data efficiency and scalability for complex interfaces and heterogeneous environments. This work establishes the GPR-ANN method as a powerful and scalable framework for constructing high-fidelity machine learning interatomic potentials, offering the computational and memory efficiency critical for the large-scale simulations needed for the simulation of materials interfaces.

In Won Yeu Annika Stuke Jon L. pez-Zorrilla James M. Stevenson David R. Reichman Richard A. Friesner Alexander Urban Nongnuch Artrith

The energy estimation procedures employed by different groups, for determining the energy of the primary $\gamma$-ray using a single atmospheric Cherenkov imaging telescope, include methods like polynomial fitting in SIZE and DISTANCE, general least square fitting and look-up table based interpolation. A novel energy reconstruction procedure, based on the utilization of Artificial Neural Network (ANN), has been developed for the TACTIC atmospheric Cherenkov imaging telescope. The procedure uses a 3:30:1 ANN configuration with resilient backpropagation algorithm to estimate the energy of a $\gamma$-ray like event on the basis of its image SIZE, DISTANCE and zenith angle. The new ANN-based energy reconstruction method, apart from yielding an energy resolution of $\sim$ 26%, which is comparable to that of other single imaging telescopes, has the added advantage that it considers zenith angle dependence as well. Details of the ANN-based energy estimation procedure along with its comparative performance with other conventional energy reconstruction methods are presented in the paper and the results indicate that amongst all the methods considered in this work, ANN method yields the best results. The performance of the ANN-based energy reconstruction has also been validated by determining the energy spectrum of the Crab Nebula in the energy range 1-16 TeV, as measured by the TACTIC telescope.

V. K. Dhar A. K. Tickoo M. K. Koul R. C. Rannot K. K. Yadav P. Chandra B. P. Dubey R. Koul

Every ring extension of $A$ by $R$ induces a pair of group homomorphisms $\mathcal{L}^{*}:R\to End_\Z(A)/L(A);\mathcal{R}^{*}:R\to End_\Z(A)/R(A),$ preserving multiplication, satisfying some certain conditions. A such 4-tuple $(R,A,\mathcal{L}^{*},\mathcal{R}^{*})$ is called a ring pre-extension. Each ring pre-extension induces a $R$-bimodule structure on bicenter $K_A$ of ring $A,$ and induces an obstruction $k,$ which is a 3-cocycle of $\Z$-algebra $R,$ with coefficients in $R$-bimodule $K_A$ in the sense of Shukla. Each obstruction $k$ in this sense induces a structure of a regular Ann-category of type $(R,K_A).$ This result gives us the first application of Ann-category in extension problems of algebraic structures, as well as in cohomology theories.

Nguyen Tien Quang Nguyen Thu Thuy

I prefer taking off this paper for the moment because of a mistake in the lemma 2.1 of the secund version. Precisely, in the proof of this lemma, it is not clear that the morphism $r\_j$ is flat, that I claim it.

Anne Moreau

Artificial neural networks (ANN) have different applications in Astronomy, including data reduction and data mining. In this work we propose the use ANNs in the identification of stellar model solutions. We illustrate this method, by applying an ANN to the 0.8M$_\odot$ star CG Cyg B. Our ANN was trained using 60,000 different 0.8M$_\odot$ stellar models. With this approach we identify the models which reproduce CG Cyg B's position in the HR diagram. We observe a correlation between the model's initial metal and helium abundance which, in most cases, does not agree with a helium to metal enrichment ratio $\Delta$Y/$\Delta$Z=2. Moreover, we identify a correlation between the model's initial helium/metal abundance and both its age and mixing-length parameter. Additionally, every model found has a mixing-length parameter below 1.3. This means that CG Cyg B's mixing-length parameter is clearly smaller than the solar one. From this study we conclude that ANNs are well suited to deal with the degeneracy of model solutions of solar type stars.

F. J. G. Pinheiro T. Simas J. Fernandes R. Ribeiro

Enhancing the robustness and accuracy of time series forecasting models is an active area of research. Recently, Artificial Neural Networks (ANNs) have found extensive applications in many practical forecasting problems. However, the standard backpropagation ANN training algorithm has some critical issues, e.g. it has a slow convergence rate and often converges to a local minimum, the complex pattern of error surfaces, lack of proper training parameters selection methods, etc. To overcome these drawbacks, various improved training methods have been developed in literature; but, still none of them can be guaranteed as the best for all problems. In this paper, we propose a novel weighted ensemble scheme which intelligently combines multiple training algorithms to increase the ANN forecast accuracies. The weight for each training algorithm is determined from the performance of the corresponding ANN model on the validation dataset. Experimental results on four important time series depicts that our proposed technique reduces the mentioned shortcomings of individual ANN training algorithms to a great extent. Also it achieves significantly better forecast accuracies than two other popular statistical models.

Ratnadip Adhikari R. K. Agrawal

The annihilator graph $AG(R)$ of the commutative ring $R$ is an undirected graph with vertex set as the set of all non-zero zero divisors of $R$, and two distinct vertices $x$ and $y$ are adjacent if and only if $ann(xy) \neq ann(x) \cup ann(y)$. In this paper, we study the embedding of the line graph of $AG(R)$ into orientable or non-orientable surfaces. We completely characterize all the finite commutative rings such that the line graph of $AG(R)$ is of genus or crosscap at most two. We also obtain the inner vertex number of $L(AG(R))$. Finally, we classify all the finite rings such that the book thickness of $L(AG(R))$ is at most four.

Mohd Shariq Praveen Mathil Mohd Nazim Jitender Kumar

In this article, we introduce a generalization of the concept of graded $r$-ideals in graded commutative rings with nonzero unity. Let $G$ be a group, $R$ be a $G$-graded commutative ring with nonzero unity and $GI(R)$ be the set of all graded ideals of $R$. Suppose that $\phi: GI(R)\rightarrow GI(R)\bigcup\{\emptyset\}$ is a function. A proper graded ideal $P$ of $R$ is called a graded $\phi$-$r$-ideal of $R$ if whenever $x, y$ are homogeneous elements of $R$ such that $xy\in P-\phi(P)$ and $Ann(x) =\{0\}$, then $y\in P$. Several properties of graded $\phi$-$r$-ideals have been examined.

Rashid Abu-Dawwas Malik Bataineh Ghida'a Al-Qura'an