Loading...
Search for: soleymani-baghshahi--mahdih
0.229 seconds

    Brain Inspired Meta Reinforcement Learning Using Brain-Inspired Networks

    , M.Sc. Thesis Sharif University of Technology Razavi Rohani, Roozbeh (Author) ; Soleymani Baghshahi, Mahdih (Supervisor)
    Abstract
    Reinforcement learning is one of the most well-known learning paradigms in biological agents and one of the most used ones for solving plenty of problems. One of the reasons for this widespread use is the low demand for supervising signals. However, the sparsity of the reward signal causes increasing in sample complexity that needs for learning new tasks. This issue makes trouble in multi-task settings, specifically.One of the most promising approaches to learning new tasks by limited interaction with the environment is meta reinforcement learning. An approach in which fast adaption becomes possible by limiting hypothesis space and creating inductive biases by learning meta parameters.... 

    Single-Cell RNA-seq Dropout Imputation and Noise Reduction by Machine Learning

    , M.Sc. Thesis Sharif University of Technology Moinfar, Amir Ali (Author) ; Soleymani Baghshah, Mahdih (Supervisor) ; Sharifi Zarchi, Ali (Supervisor) ; Goodarzi, Hani (Co-Supervisor)
    Abstract
    Single-cell RNA sequencing (scRNA-seq) technologies have empowered us to study gene expressions at the single-cell resolution. These technologies are developed based on barcoding of single cells and sequencing of transcriptome using next-generation sequencing technologies. Achieving this single-cell resolution is specially important when the target population is complex or heterogeneous, which is the case for most biological samples, including tissue samples and tumor biopsies.Single-cell technologies suffer from high amounts of noise and missing values, generally known as dropouts. This complexity can affect a number of key downstream analyses such as differential expression analysis,... 

    Effects of nucleation agents on the preparation of transparent glass-ceramics

    , Article Journal of the European Ceramic Society ; Volume 32, Issue 11 , August , 2012 , Pages 2989-2994 ; 09552219 (ISSN) Ghasemzadeh, M ; Nemati, A ; Baghshahi, S ; Sharif University of Technology
    Elsevier  2012
    Abstract
    Formation of transparent glass-ceramic in the system MgO-SiO 2-Al 2O 3-K 2O-B 2O 3-F with and without addition of LiF and NaF has been investigated. Crystallization of glass-sample was conducted by controlled thermal heat-treatment, at determined nucleation and crystallization temperatures. In this regard, the effects of addition of LiF and NaF were investigated on the crystallization behavior and transparency of the samples. Low transmission (less than 80% at 600. nm) was observed in the basic composition (K).The addition of NaF and LiF caused more intense phase separation in the system. The results indicated that the glass-ceramic can remain transparent if fine grains with nano size are... 

    The relation between particle size and transformation temperature of gibbsite to αLPHA-alumina

    , Article Mineral Processing and Extractive Metallurgy: Transactions of the Institute of Mining and Metallurgy ; Volume 131, Issue 2 , 2022 , Pages 111-121 ; 25726641 (ISSN) Ahmadabadi, M.N ; Nemati, A ; Arzani, K ; Baghshahi, S ; Sharif University of Technology
    Taylor and Francis Ltd  2022
    Abstract
    The transformation of gibbsite to α-alumina occurs in the range 1100 to 1300°C. The higher the calcination temperature, the larger the crystallites and this leads to abnormal grain growth. The main goal of this research was to reduce the transformation temperature of gibbsite to α-alumina by reducing the gibbsite particles size. The sodium content of the gibbsite used in the study was reduced to less than 0.1% by washing it in HCl solution. It was then ground in a fast mill at ambient conditions for different times then calcined at different temperatures. The microstructure, particles size and thermal behaviour of the samples were examined with SEM, XRD, XRF, PSA and STA, respectively. In a... 

    Synthesis and characterization of sol-gel derived hydroxyapatite-bioglass composite nanopowders for biomedical applications

    , Article Journal of Biomimetics, Biomaterials, and Tissue Engineering ; Volume 12, Issue 1 , 2012 , Pages 51-57 ; 16621018 (ISSN) Adibnia, S ; Nemati, A ; Fathi, M. H ; Baghshahi, S ; Sharif University of Technology
    2012
    Abstract
    The main purpose of this study is to prepare and characterize hydroxyapatite (HA)-10%wt bioglass (BG) composite nanopowders and its bioactivity. Composites of hydroxyapatite with synthesized bioglass are prepared at various temperatures. Suitable calcination temperature is chosen by evaluating of the phase composition. X-ray diffraction (XRD), Transmission electron microscopy (TEM) and Scanning electron microscopy (SEM) techniques are utilized to characterize the prepared nanopowders. The bioactivity of the prepared composite samples is evaluated in an in vitro study by immersion of samples in simulated body fluid (SBF) for predicted time. Fourier transformed infrared (FTIR) spectroscopy and... 

    The relation between particle size and transformation temperature of gibbsite to αLPHA-alumina

    , Article Mineral Processing and Extractive Metallurgy: Transactions of the Institute of Mining and Metallurgy ; 2020 Ahmadabadi, M. N ; Nemati, A ; Arzani, K ; Baghshahi, S ; Sharif University of Technology
    Taylor and Francis Ltd  2020
    Abstract
    The transformation of gibbsite to α-alumina occurs in the range 1100 to 1300°C. The higher the calcination temperature, the larger the crystallites and this leads to abnormal grain growth. The main goal of this research was to reduce the transformation temperature of gibbsite to α-alumina by reducing the gibbsite particles size. The sodium content of the gibbsite used in the study was reduced to less than 0.1% by washing it in HCl solution. It was then ground in a fast mill at ambient conditions for different times then calcined at different temperatures. The microstructure, particles size and thermal behaviour of the samples were examined with SEM, XRD, XRF, PSA and STA, respectively. In a... 

    The effect of mixing molar ratios and sand particles on microstructure and mechanical properties of metakaolin-based geopolymers

    , Article Materials Chemistry and Physics ; Volume 240 , 2020 Riahi, S ; Nemati, A ; Khodabandeh, A. R ; Baghshahi, S ; Sharif University of Technology
    Elsevier Ltd  2020
    Abstract
    In this study the influence of the molar ratios such as SiO2:Al2O3:Na2O:H2O, as well as the sand particles on the mechanical performance, shrinkage and microstructure of metakaolin based geopolymers was studied. Considering different content of the reactive silica and alumina in metakaolin, forty eight different compositions were prepared. The proper condition for achieving the highest mechanical performance as well as less structural defects by tailoring the curing condition, molar ratios and addition of sand particles are presented. Special attention was paid to the incorporation of sand particles up to 70 wt% on the microstructure, shrinkage and mechanical properties of metakaolin based... 

    Investigation of interfacial and mechanical properties of alumina-coated steel fiber reinforced geopolymer composites

    , Article Construction and Building Materials ; Volume 288 , 2021 ; 09500618 (ISSN) Riahi, S ; Nemati, A ; Khodabandeh, A. R ; Baghshahi, S ; Sharif University of Technology
    Elsevier Ltd  2021
    Abstract
    In this study, alumina coated steel reinforced geopolymer composites were successfully fabricated and their interfacial and mechanical properties were investigated. The fiber–matrix interface characteristics were assessed using single-fiber push-out and water contact angle experiments as well as SEM-EDS analysis. Finally, the role of alumina coating on the compressive, flexural and toughness of the geopolymer composites was investigated. The morphological studies revealed that alumina coating on steel can chemically bonded and microstructurally integrated with the surrounding matrix. The push-out test showed that the interfacial shear strength was increased approximately 150% in composites... 

    Influence of NaF on crystallization and machinability of mica glass ceramics

    , Article Synthesis and Reactivity in Inorganic, Metal-Organic and Nano-Metal Chemistry ; Volume 42, Issue 7 , May , 2012 , Pages 958-964 ; 15533174 (ISSN) Ghasemzadeh, M ; Nemati, A ; Hamnabard, Z ; Baghshahi, S ; Golikand, A. N ; Sharif University of Technology
    T&F  2012
    Abstract
    Effect of NaF on crystallization kinetics, microstructure, and mechanical properties of mica glass ceramics were investigated by the differential thermal analysis (DTA), X-ray diffractometry (XRD), scanning electron microscopy (SEM), and microhardness tests. Non-isothermal DTA experiments showed that the crystallization activation energies of base glasses are changed in the range of 235-246 KJ/mol, while the crystallization activation energies of samples with addition of NaF are changed in the range of 263-367 KJ/mol. The increase of crystallization temperature is helpful for the increase of aspect ratio, and the microstructure of the glass ceramics becomes interconnected, which contributes... 

    Utilization of DTA in the determination of a crystallization mechanism in transparent glass-ceramics with a nanocrystalline structure

    , Article Synthesis and Reactivity in Inorganic, Metal-Organic and Nano-Metal Chemistry ; Volume 41, Issue 5 , 2011 , Pages 561-570 ; 15533174 (ISSN) Ghasemzadeh, M ; Nemati, A ; Golikand, A. N ; Hamnabard, Z ; Baghshahi, S ; Sharif University of Technology
    2011
    Abstract
    Non-isothermal differential thermal analysis (DTA), X-ray diffraction (XRD), and scanning electron microscopy (SEM) were used to study the nucleation and crystallization behavior of mica glass-ceramics with LiF as nucleating agent. The models enabled establishing the kinetic parameters for crystal growth of individual phases. The activation energies for crystal growth were found to be in the range of 161-301 KJ/mol, 416-424 KJ/mol, and 583-1011 KJ/mol for base glasses, samples with substitution of Li2O for K2O and samples with addition of LiF, respectively. Formation of transparent glass-ceramics from studied glass-samples has been investigated. Transparency is assumed to occur in the... 

    Sliding mode leader following control for autonomous air robots

    , Article 2011 IEEE/SICE International Symposium on System Integration, SII 2011, 20 December 2011 through 22 December 2011 ; December , 2011 , Pages 972-977 ; 9781457715235 (ISBN) Soleymani, T ; Saghafi, F ; Sharif University of Technology
    2011
    Abstract
    In this paper, we propose a leader following control for autonomous air robots. The separated design strategy with kinematic acceleration commands is used. The location of the robot with respect to the leader is specified by a range and two angles. We obtain the kinematic model of the system represented by the state-space equations. The controller is designed based on the sliding mode control which asymptotically stabilizes the tracking errors in presence of uncertainties and disturbances. In order to implement the leader following controller in the air robots, a control system is introduced which converts the acceleration commands to the actuator commands. Simulations are provided to show... 

    Behavior-based acceleration commanded formation flight control

    , Article ICCAS 2010 - International Conference on Control, Automation and Systems 2010, Article number 5670304, Pages 1340-1345 ; 2010 , Pages 1340-1345 ; 9781424474530 (ISBN) Soleymani, T ; Saghafi, F ; Sharif University of Technology
    2010
    Abstract
    In this paper, the design of a formation flight controller is investigated. Each vehicle in the formation is controlled by designing two separate control loops. The formation flight controller placed in the outer loop employs behavior-based control as a distributed control strategy to steer the vehicle by producing acceleration commands and the control system placed in the inner loop is to convert these commands to the actuator commands. Leader following architecture is applied to define the structure for the formation flight. To study the pragmatic issues of the proposed formation flight controller, it is implemented into multiple micro air vehicles which are modeled by a... 

    Fuzzy trajectory tracking control of an autonomous air vehicle

    , Article ICMEE 2010 - 2010 2nd International Conference on Mechanical and Electronics Engineering, Proceedings, 1 August 2010 through 3 August 2010 ; Volume 2 , August , 2010 , Pages V2347-V2352 ; 9781424474806 (ISBN) Soleymani, T ; Saghafi, F ; Sharif University of Technology
    2010
    Abstract
    The development and the implementation of a new guidance law are addressed for a six dimensional trajectory tracking problem, three dimensions for position tracking and three dimensions for velocity tracking, of a micro air vehicle. To generate the desired trajectory a virtual leader is defined which is moved in space. In the guidance law, position and velocity feedbacks are used by fuzzy controllers to generate two acceleration commands. Then, a fuzzy coordinator is applied to coordinate the acceleration commands. Nonlinear six-degree-of-freedom equations of motion are used to model the vehicle dynamics. Also, a bank-to-turn acceleration autopilot for vehicle is considered to follow the... 

    Unsupervised learning for distribution grid line outage and electricity theft identification

    , Article 2019 Smart Gird Conference, SGC 2019, 18 December 2019 through 19 December 2019 ; 2019 ; 9781728158945 (ISBN) Soleymani, M ; Safdarian, A ; Sharif University of Technology
    Institute of Electrical and Electronics Engineers Inc  2019
    Abstract
    The development of smart meters enables situational awareness in electric power distribution systems. The situational awareness provides significant advantages such as line outage and electricity theft detection. This paper aims at using smart meter data to detect these anomalies. To do so, an appropriate cluster-based method as an unsupervised machine learning approach is applied. A stochastic method based on conditional correlation is also proposed to localize the anomalies. It is shown that this can be done by detecting changes in bus connections using present and historical smart meter data. Therefore, network topology inspection can be avoided if the proposed method is applied. A... 

    Development of a Distributed Algorithm for Flocking of Non-Holonomic Aerial Agents

    , M.Sc. Thesis Sharif University of Technology Soleymani, Touraj (Author) ; Saghafi, Fariborz (Supervisor)
    Abstract
    The goal of this project is the development of a control algorithm for a flock of non-holonomic aerial agents. For this purpose,the swarm architecture having some unique features such as robustness, flexibility, and scalability is utilized. Swarm is defined as a group of simple agents having local interactions between themselves and the environmentwhich shows an unpredictable emergent behavior.Behavior based control which is inspired from the animal behaviors is employed to control the swarm of mobile agents. Accordingly, the necessary behaviors which are distance adjustment, velocity agreement, and virtual leader tracking together with a fuzzy coordinator are designed. In this study, in... 

    Deep Zero-shot Learning

    , M.Sc. Thesis Sharif University of Technology Shojaee, Mohsen (Author) ; Soleymani, Mahdieh (Supervisor)
    Abstract
    In some of object recognition problems, labeled data may not be available for all categories. Zero-shot learning utilizes auxiliary information (also called signatures) describing each category in order to find a classifier that can recognize samples from categories with no labeled instance. On the other hand, with recent advances made by deep neural networks in computer vision, a rich representation can be obtained from images that discriminates different categorizes and therefore obtaining a unsupervised information from images is made possible. However, in the previous works, little attention has been paid to using such unsupervised information for the task of zero-shot learning. In this... 

    Multi-Modal Distance Metric Learning

    , M.Sc. Thesis Sharif University of Technology Roostaiyan, Mahdi (Author) ; Soleymani, Mahdieh (Supervisor)
    Abstract
    In many real-world applications, data contain multiple input channels (e.g., web pages include text, images and etc). In these cases, supervisory information may also be available in the form of distance constraints such as similar and dissimilar pairs from user feedbacks. Distance metric learning in these environments can be used for different goals such as retrieval and recommendation. In this research, we used from dual-wing harmoniums to combining text and image modals to a unified latent space when similar-dissimilar pairs are available. Euclidean distance of data represented in this latent space used as a distance metric. In this thesis, we extend the dual-wing harmoniums for... 

    Unsupervised Domain Adaptation via Representation Learning

    , M.Sc. Thesis Sharif University of Technology Gheisary, Marzieh (Author) ; Soleymani, Mahdieh (Supervisor)
    Abstract
    The existing learning methods usually assume that training and test data follow the same distribution, while this is not always true. Thus, in many cases the performance of these learning methods on the test data will be severely degraded. We often have sufficient labeled training data from a source domain but wish to learn a classifier which performs well on a target domain with a different distribution and no labeled training data. In this thesis, we study the problem of unsupervised domain adaptation, where no labeled data in the target domain is available. We propose a framework which finds a new representation for both the source and the target domain in which the distance between these... 

    Deep Learning for Multimodal Data

    , M.Sc. Thesis Sharif University of Technology Rastegar, Sarah (Author) ; Soleymani, Mahdieh (Supervisor)
    Abstract
    Recent advances in data recording has lead to different modalities like text, image, audio and video. Images are annotated and audio accompanies video. Because of distinct modality statistical properties, shallow methods have been unsuccessful in finding a shared representation which maintains the most information about different modalities. Recently, deep networks have been used for extracting high-level representations for multimodal data. In previous methods, for each modality, one modality-specific network was learned. Thus, high-level representations for different modalities were extracted. Since these high-level representations have less difference than raw modalities, a shared... 

    Data Mining of Smart Metering Data for Abnormality Detection in Electric Energy Consumption

    , M.Sc. Thesis Sharif University of Technology Soleymani, Mohammad (Author) ; Safdarian, Amir (Supervisor)
    Abstract
    The development of smart meters enables gathering and analysis of a large amount of data about electrical energy consumption in electric power distribution systems. This data and the obtained behavioral patterns of customers have a wide variety of applications. To name a few, classification of customers based on their consumption patterns, damaged smart meter identification, non-technical loss identification and measuring participation rate of customers in demand response programs are among the applications. So far, many studies have been done for consumption pattern identification. However, abnormality detection in electric energy consumption has captured growing attention due to the...