Sharif Digital Repository

Probabilistic graphical models have great applications in studying and analyzing realworld data. For instance, these models have been used in reconstructing gene regularity networks. Specifically, learning the edges’ structure of graphical models is of great importance.Knowledge about the underlying structure of a graphical model brings about a valuable framework for the decomposition of the model’s distribution and reveals important information such as dependency among dimensions of samples, etc. Most existing methods for structure learning obtain the underlying structure of the model in a centralized fashion and without considering noise in data. In many applications, data exist in a... 

In this study, we aim to learn representations from microscopic cell images that effectively capture the features of drugs affecting the cells, allowing us to identify effective drugs for treating a disease. We employ two parallel learning paths using predictive and generative models. Specifically, we have achieved a predictive model on the RxRx19a dataset that, unlike previous models, is interpretable, optimized, and robust to dras- tic changes in drug properties. Additionally, we have developed the first generative model on this dataset, which not only generates high-quality images but also discovers a meaningful latent space. This latent space divides the representation into relevant... 

Around world and the networks within it can be modeled in various templates. Graph structure is one of those templates in which objects and relations may have more than one types. We call this phenomenon "heterogeneity".Heterogeneity makes the networks hard to model and that is why the proposed methods for modeling the networks assumed the network structures homogeneous. This assumption may cause data loss due to ignoring the variety of types in network objects and relations and this loss can lessen the accuracy of data mining tasks.To tackle the challenge of data loss in the mentioned assumption, learning representations for heterogeneous information networks (HINs) was introduced. HINs... 

Representation learning methods on graphs have enabled using machine learning methods on graphs' discrete structure by transferring them to a continuous domain. As graphs' structures are not always static and may evolve through time, dynamic representation learning methods have recently gained scholars' attention. Several methods have been proposed to enable the model to update the embeddings graph changes, or new interactions happen between nodes. These online methods could significantly reduce the learning time by refreshing the model as the changes occur, so we don't need to retrain the model with the complete graph information. Moreover, by using the temporal information of interactions,... 

Representation learning refers to mapping the input data to another space, usually with lower dimensions than the input space. This task can be helpful in improving the performance of methods in downstream tasks, compression, and improving sample generation in generative models. Representation learning is a problem connected to information theory, and information theory's concepts and quantities are used widely in representation learning models. Besides, the representation learning problem is closely related to latent variable generative models. These models usually learn useful representations in their process of training, implicitly or explicitly. So, the usage of latent variable... 

Since, presence of sulfur compounds in fuels leads to environmental pollution and deacrese of fuel quality, fuel desulfurization should be considered for enhancing the fuel charachteristics. Electrochemical desulfurization was recogonized as one of desulfurization techniques that uses anode and cathode in an electrolytic environment under mild temperature and pressure. In this research, electrochemical desulfurization of crude gas-oil has been studied over electrodes of copper and stainless steel in the presence of NaOH and sulfuric acid. In this research, the effects of operational prameters like stirring rate, temperature, applied potential, sodium hydroxide/sulfuric acid addition as... 

Oxidative desulfurization (ODS) is one of the desulfurization processes that posseses several important advantages such as mild operating conditions. In the present work, oxidative desulfurization of kerosene with a total sulfur content of 2335 ppm and gas oil with a total sulfur content of 7990 ppm were carefully investigated. By evaluation of different oxidation systems, the hydrogen peroxide-formic acid system was selected for the oxidation of sulfur-containing compounds of the light fuel oils. The effects of main operating parameters of the oxidation system including temperature, hydrogen peroxide to sulfur molar ratio (O/S), and formic acid to sulfur molar ratio (Acid/S) on the... 

In this study, the pervaporation method was used for desulfurization of gasoline. Thiophene was used as a sulfur agent and n-heptane as a representative of hydrocarbon compounds in gasoline. The composite membranes of the active PDMS layer containing the UiO-66 metal-organic framework nanoparticles are based on porous membrane PVDF. The effect of adding PVP polymer on the porosity of the PVDF membrane has been investigated. Scanning electron microscopy (SEM), Fourier transform infrared spectrum (FTIR) and X-ray diffraction (XRD) have been used to characterize nanoparticles and synthesized membranes. The effect of adding nanoparticles on the membrane performance and its sowelling effect in... 

Constant drilling in the shale formation causes many problems such as shale swelling and well instability. Annually many costs are spent dealing with these problems in the drilling industry. Many factors can affect this issue. The most important factor is the type of drilling fluid and its additives. The use of oil-based mud is the best option to deal with the clays instability but, due to environmental problems resulting from OBM, their use has been banned in most parts of the world. Therefore, the water-based mud is the only option for the oil industry that can modify its property by use of additives. In this study, the muds contain Nano silica have been introduced. Silicate as shale... 

One of the ultimate goals in robotics is to design autonomous robots. Among other things, this means a robot has to be able to plan its own motion. To be able to plan a motion, a robot must have some knowledge about the environment in which it is moving. Some of this information can be provided by a floor plan. For other information the robot will have to rely on its sensors. In this thesis, we consider that the environment is unknown for the robot. So, the robot uses weak sensor to planning its motion. The sensor can distinguish between colors. The robot can follow the colors. Therefore, we need to specify the umber of colors in which the robot can navigate by them. We propose an efficient... 

We consider extending the visibility polygon of a given point q inside a simple polygon by converting one or some of the edges to mirrors. It is known that the visibility polygon of a point inside a polygon in the presence of a mirror can be computed in linear time. We propose a linear time algorithm that checks whether a given segment inside a simple polygon is visible by a given point q via a mirror-edge, and finds all such edges. And we prove that finding edges to add exactly k unit squares to the visibility polygon of a query point q in most cases is NP-complete and its minimization is NP-hard. These cases contain single and muliple, either with secular or difuse reflections. Moreover,... 

Many natural and engineering materials have a heterogeneous structure at a certain level of observation. These materials are often referred to as composite materials or multi-phase materials or heterogeneous materials. It has been widely recognized that many macroscopic phenomena originate from the mechanics of the microstructural constituents, such as inclusions, cracks, voids, etc. The size, shape, spatial distribution, volume fraction and properties of the microstructural constituents have a significant impact on the behavior of the material observed at the macroscale. The nature of hydrocarbon reservoirs as multi-phase porous media are known for heterogeneous media at various multiple... 

This thesis is a survey of results of studies and researches in theory of Division rings and multiplicative groups of division rings and relations between group strcture and algebraic structure of division rings.in this case,we emphesized on study of finitely generated,maximal,nilpotent and soluble subgroups of division rings.we also,study the Valuation theory on division rings and Reduced K-theory of division rings and relations between these theories and group structure of division rings,morever,those which is finite dimensional over their center as vector spaces.at the end,we shortly,study divisible division rings and radicable division rings  

Our bodies are made from billions of individual cells and DNA is the control central of each cell. In addition to linear DNAs, there are circular DNAs in nature. Circular DNAs cause new topological structures such as knots and catenanes. These topological structures form during biological proccesses such as replication. For example during replication, two daughter circular DNAs may link together and form catenanes. Anzymes named topoisomerases can simplify the topology of circular DNAs. Type 2 topoisomerases is used for simplification of catenanes. These anzymes do it with breaking a double-strand DNA and passing through another one. In comparison with DNAs, topoisomerases are very small and... 

AdS/CFT correspondence is a suitable framework for studying strongly coupled gauge theories. It corresponds a strongly coupled gauge theory to a classical gravity theory with an extra dimension. Quark-gluon plasma which is produced in accelerators is strongly coupled. We use AdS/CFT correspondence to calculate Meson thermalization time by approximating the collisions by a sudden change of baryon number chemical potential. On the gravity side, this is followed with an apparent horizon formation on probe-brane which is the signal of thermalization  

Capturing the atmospheric CO2 and turning it into value added products is a topic that has been extensively studied in recent years. One of the methods through which CO2 is valorized is via converting it into carbon nanofiber, CNFs. CNFs can be an important material in future human activities due to its valuable properties such as wide range of electrical conductivity, their superior flexibility and strength, But it is currently obstructed in use due to the high cost of manufacturing and the complexity of the synthesis. Furthermore, in today's world, human life is heavily influenced by greenhouse gases, and avoiding emitting gases like CO2, is almost infeasible. This thesis provides an... 

It is generally believed that the only utility of tests as a critical instrument is to be at the service of educational institutions. To broadly consider their different usefulness, tests can serve other purposes like focusing on the ethical and moral issues. As it is called Critical Language Testing, tests are not reliable instruments and they have the capacity of being used not only in educational, but also in political contexts. This philosophy has not been brought into perspective, and, as a result, the literature deserves more comprehensive attention. This study, guided by the principles of Grounded Theory and Qualitative research design, pursued three purposes: To investigate the... 

In this thesis, the aim is to stating some of the noncommutative geometry(NCG) applications in formulating and describing the gravity. So, in the first chapter, we will work on some of metric aspects of NCG. In this chapter, we will review elements of the spin geometry, specifically we will define the Dirac operator which is the most important object in this thesis. Then, we will try to find counterparts of some of Riemannian geometric notions like integral and distance in the NCG framework. We will define some notions such as noncommutative infinitesimals, Dixmier trace, Wodzicki residue, noncommutative integral and spectral triples. Then, in the second chapter, we will work on some of the... 

The observations of the cosmic background radiation and large-scale structures show that the standard cosmological model is in good agreement with observational data. The standard model is based on the cosmological constant, dark matter, and Gaussian, nearly scale-invariant, adiabatic, and isotropic initial conditions. Although this model has been phenomenologically successful, but the problems of dark energy, dark matter, missing baryons and early universe remain unresolved. One of the most important theoretical questions in cosmology is whether the cosmological constant is the reason for accelerated cosmic expansion. In other words, is Einsteinian gravity the correct theory on the cosmic... 

Let (P;≼) be a partially ordered set (poset, briefly) with a least element 0. In this thesis, we deal with zero-divisor graphs of posets. We show that if the chromatic number r(P) and the clique number r(P) (x(r(P)) and !(r(P)), respectively) are finite, then x(r(P)) = !(w(P)) = n in which n is the number of minimal prime ideals of P. We also prove that the diameter of such a graph is either 1, 2 or 3 while its girth is either 3, 4 or 1