Sharif Digital Repository

In general, Constrain Satisfaction Problems (CSP’s) are NP-complete, however, this is not the case, if we impose some restrictions on the constraints. Due to importance and applications of CSP’s in different areas of engineering and science including artificial intelligence, relational databases and logic, investigating the computational complexity of CSP’s is an important issue. Computational complexity of CSP’s has been investigated from a variety of aspects, including the Universal Algebraic approach that has given rise to a solution of dichotomy conjecture recently. In this thesis based on contributions of Bulatov, Krokhin and Jeavons, we try to explain this universal algebraic approach... 

Two new frameworks for achievability proof in network information theory (NIT) are introduced . The first framework which we call it OSRB (Output Statistics of Random Binning) framework, uses random binning arguments and is based on a duality between channel and source coding problems. Furthermore, the OSRB framework uses pmf approximation arguments instead of counting and typicality which are used widely in conventional achievability proofs in NIT. This allows for proving coordination and strong secrecy problems, where certain statistical conditions on the distribution of random variables need to be satisfied. These statistical conditions include independence between messages and... 

Federated learning (FL) systems, recognized for their decentralized machine learning capabilities across multiple nodes without aggregating data, often grapple with high communication overhead and stringent privacy requirements. This thesis proposes an innovative approach that utilizes random projection and compressed sensing techniques to project gradients in weight spaces into a lower-dimensional subspace, effectively reducing communication overhead to O(logd) bits. This method distinguishes itself by addressing both communication efficiency and privacy simultaneously, a notable advancement over traditional methods that typically focus on one at the expense of the other. By implementing... 

Based on the importance of DNA/RNA binding proteins in different cellular processes, finding binding sites of them play crucial role in many applications, like designing drug/vaccine, designing protein, and cancer control. Many studies target this issue and try to improve the prediction accuracy with three strategies: complex neural-network structures, various types of inputs, and ML methods to extract input features. But due to the growing volume of sequences, these methods face serious processing challenges. So, this paper presents KDeep, based on CNN-LSTM and the primary form of DNA/RNA sequences as input. As the key feature improving the prediction accuracy, we propose a new encoding... 

One of the most practical deformation processes has been used to produce profiles with a diversity of shapes is forward extrusion. Minimizing the extrusion pressure along with reducing the curvature of the final product have always been the main concerns of the researchers in this area. To achieve the former, researchers have designed nonlinear dies and proposed various equations to determine proper die profile. To accomplish the latter, they have used bearing at the exit section of the extrusion die. The primary object of the current study is to design a proper bearing in order to eliminate the curvature of the final product in extrusion process. The effects of both bearing and die profile... 

Speech synthesis and its applications have been very attractive recently. The main purpose of this technique is to produce a speech signal with natural characteristics of human speech like prosody and emotion. Among all existing methods for speech synthesis, statistical parametric speech synthesis methods are more promising because ofhigher flexibility in comparison to other methods. One of the applications of speech synthesis is speech to speech translation. In these systems, the generated voice in target language should have the same characteristics as the input voice in source language. The main purpose of this research is to review and evaluate the cross lingual speaker adaptation... 

Surface piercing propellers are particular type of supercavitating propellers that are commonly used for High-speed vessels. Most studies on this type of propellers has been investigating the hydrodynamic forces. But in recent years with increase in SPPs diameter used in vessels, structural analysis of this type of propellers is considered. For this purpose, studies on the stresses exerted on the propeller structures under load is done with the help of Hydro elastic methods. In this type of analysis, structura of propeller is intended to be flexible and Displacements under pressure checked and Tensions resulting from it are studied.
The present Thesis using ANSYS software to analyze a... 

Pricing plays an important and essential role in the profit and income of companies. The importance of pricing is not only related to its role in the company's profitability, but it also changes the customer's understanding and loyalty towards the company and can create the company's reputation or destroy it. Determining the right price will increase product sales and increase customer loyalty and create a competitive advantage for the company. One of the most important and influential variables in product pricing is the amount of demand. The main challenge of companies for product pricing is the uncertainty in their demand. In order to deal with this problem, data-driven pricing is used.... 

The increasing demand for autonomous vehicles and reliable communication protocols and hardware interfaces, such as CAN bus and USB, underscores the necessity for stable clock sources that maintain a low temperature coefficient (TC) over wide temperature ranges. This demand is particularly emphasized in applications such as wearables, network sensors, downhole devices, WSNs, and IoT, where long-lasting battery life and frequency-stable clock sources over a broad temperature range (e.g. -20 °C to 100 °C) are crucial. Traditionally, variations in frequency caused by temperature have been mitigated by employing off-chip components like crystals or ceramic based oscillators, but this approach... 

The increasing demand for communication bandwidth and limited spectrum availability have heightened the complexity of radio front-end circuits in IoT applications. Achieving spectral efficiency is a key challenge, particularly for IoT devices operating at specific frequency bands such as 315 MHz, 433 MHz, 868 MHz, and 915 MHz.Due to the limited linearity of the transmit path, harmonic distortion components, known as counter intermodulation (CIM) products, are generated. These CIM products can directly fall into the receiver (RX) band or enter it via cross-modulation, thereby degrading the frequency division performance. Additionally, CIM products can interfere with protected bands, violating... 

Undoubtedly, machine learning and its practical applications have become essential topics in research and industry. Among these, representation learning holds special importance. The goal in this field is to find low-dimensional and meaningful representations of high-dimensional data that encapsulate essential information. Most existing methods focus only on the statistical correlations in data, which do not necessarily indicate real relationships between variables and can affect learning performance. This becomes even more problematic when the distribution of test data differs from the training data, as it makes the learned representations less generalizable. To tackle this problem, causal...