Abstract

With the booming online market, managing the warehouse inventory is one of the most essential and challenging tasks. The management of inventory will be more efficient if it is automated using robots. The robots can work faster than humans, the robots work at a constant speed with no breaks, and do tasks in more repetition than humans like fetching inventory from warehouse. But for the robots to perform tasks like putting objects in racks, fetching objects from rack, re-organising the racks to make more space they need to have an understanding of the warehouse environment. To understand the warehouse environment the robots needs to create a 3D map of the warehouse consisting space to move for the robot as well as identify the racks and boxes so that its able to plan the execution of tasks. Towards solving this problem in this thesis we address problem of freespace estimation for rack shelves. Given a monocular RGB image captured from a camera mounted on a robotic arm. We aim to predict the Top-view and Front-view layouts so as to create a 3D reconstruction of rack and objects present in the Monocular RGB image. We propose a simple yet effective network architecture RackLay, which takes a monocular RGB image as input and outputs the Top-view and Front-view layout of all the shelves comprising the rack visible in the image. The Network can learn two kinds of layout representations, one in the canoncial frame centered on the shelf, called the shelf-centeric layout and the other in a frame with respect to the camera, called the ego-centric layout. Apart from portraying the versatility of the network, they lend to various useful applications. Since there are very few publicly available datasets for warehouse settings, we also introduce the synthetic data generation pipeline termed as WareSynth, which can be used to generate 3D warehouse scenes, automate the process of data capture and generate corresponding annotations. WareSynth can be used for various tasks such as 2D/3D object detection, semantic/instance segmentation, layout estimation, 3D scene navigation and mapping, 3D reconstruction etc. The same pipeline can also be modified to other kind of scenes such as supermarkets, greenhouses by changing the database of objects and placement parameters hence this pipeline open gates for further research in similar environments.

Abstract

Blockchains lie at the heart of Bitcoin and other cryptocurrencies that have shown great promise to revolutionize finance and commerce. The novel applications of blockchain technology are derived from the unique combination of properties of blockchains: immutability, incorruptibility, consensus, transparency, decentralization and ordering. However, they face technical challenges when it comes to scaling to support greater demand while maintaining their desirable security properties. In an exciting line of recent work, many researchers have proposed various scalable blockchain protocols that demonstrate the potential to solve these challenges. However, many of these protocols come with the assumptions of honest majority and symmetric network access which may not accurately reflect the real world where the participants may be self-interested or rational. In this thesis, we work towards analysing blockchain protocols in settings with rational participants and use game-theory to predict their behaviour. For the first time in literature, we study blockchain protocols in a setting with asymmetric network access and highlight the role of network fairness in ensuring the security of a blockchain protocol. We demonstrate via simulations, the effect of lack of network fairness in Bitcoin and the equilibrium behaviour of the participants which results in loss of both security and performance of the protocol. In order to further highlight the importance of network fairness in designing scalable blockchain protocols, we showcase incentive-driven deviations in the OHIE blockchain protocol proposed recently and the loss of security caused by these deviations. We then study a class of blockchain protocols designed by layering two distinct types of consensus protocols and determine the parameters required to ensure game-theoretic soundness and security. We also formally study the reward schemes employed by Proof-of-Work-based blockchain protocols and showcase the lack of game-theoretic soundness. Towards this, we also propose a class of game-theoretically sound reward schemes. Finally, we extend our work by developing a model for layered-blockchain protocols and present a game-theoretic model for analyzing the security of layered-blockchain protocols. Therefore, in this thesis we present novel work that aims to guide the development of fair and game-theoretically sound blockchain protocols.

Abstract

Natural language processing (NLP) is a bridge between the computer and human interactions in their natural language. NLP has wide variety of applications such as machine translation, text summarization, question-answering, sentiment analysis, etc. All these applications have created huge impact in the society with different use cases such as chatbots, voice assistants (Alexa, Siri), recommendation systems (YouTube, Hotstar) etc. Most of these NLP applications are limited to few high resource languages like English. Notably, in an Indian scenario, where each state is having its own language, only 10% of people communicate in English. This motivates us to develop NLP systems and resources in Indian languages that will create a huge impact in multilingual Indian society. In this thesis, we created question-answering (QA) and text-summarization resources and systems in Telugu language. The main aim of QA system is to provide an accurate and concise answer to the question asked by human in natural language. In this thesis, we created a question classification dataset which consists of 1037 samples and also explained the ambiguities, challenges involved in creating the dataset. We built an end to end pipeline for QA system and named it as AVADHAN. We performed comparisons between three different classifiers for the Telugu Question Classification (QC) module. QC will be helpful to reduce the search space while extracting the answer for the given query. Text summarization is a way of obtaining short and precise summary from the given document of arbitrary length. In this thesis, we have proposed a pipeline that crowd-sources summarization data and then aggressively filters the content with automatic and partial expert evaluation. With this pipeline we have created TeSum: high quality human generated abstractive summarization corpus for Telugu. This corpus consists of 20329 high quality articlesummary pairs and this is the first high quality and large abstractive summarization corpus in Telugu as per our knowledge. We have also explored different sequence to sequence neural networks on TeSum corpus and provided ROUGE scores.

Abstract

Since the 1964 Niigata earthquake (the epicentre is located at the continental shelf off the northwest coast of Honshu, Japan), dynamic soil-structure interaction has been considered an important factor in many important structures such as tall buildings, bridges, nuclear power plants, etc. As soil-structure interaction analysis is a complex phenomenon, researchers have developed different techniques through experimental, analytical, and numerical approaches. Amongst all the techniques numerical methods are found more reliable in the design of structures to include the effects of soil-structure interaction. However, radiating waves from structure are one of the major concerns in numerical modelling of the soil-structure interaction. To solve the radiating wave propagation problems using finite element analysis (FEA), it is required that the boundary must be terminated at some finite location. This truncation of the model at the finite boundary will cause the reflection of radiating waves. The reflected waves from the boundary will affect the solution and may lead to instabilities in the numerical analysis. Therefore, it is necessary to provide an artificial boundary condition that will transmit the outwardly propagating waves with minimal or negligible reflections. The primary objective of this research is to develop an efficient radiating boundary condition for numerical simulation of wave propagation in nonlinear, unbounded spatial domains. Despite several attempts by the researchers, the challenge of developing a computationally efficient absorbing boundary condition (ABC) to resemble the Sommerfeld radiation condition has not been well addressed. Absorbing Boundary Conditions (ABC), also called Local Absorbing Boundary Conditions are simple and computationally efficient, but they produce spurious reflections when the wave impinges on the boundary in a direction other than the normal. Absorbing layer techniques are efficient in absorbing outwardly propagating wave energy, but these techniques require many layers. Researchers have also attempted to combine the Absorbing Boundary Conditions (ABC) with the Layers by Increasing Damping (ALID) to utilize the advantages of both methods. Since ABC is only applicable to wave propagation in elastic media, the attempt to combine the two techniques becomes unsuccessful due to an impedance mismatch. In this thesis, a new absorbing boundary condition for wave propagation in a viscoelastic medium (VABC) is proposed. The method is an extension of the standard ABC proposed by Lysmer and Kuhlemeyer (1969). The proposed method does not converge to Kelvin-type viscoelastic materials but can be applied to Maxwell-type viscoelastic materials, i.e., only mass proportional damping is considered. The accuracy of the method is studied for viscoelastic wave propagation problems, and the results are compared with the standard ABC and analytical solutions.  The analytical and numerical results show that the VABC boundary conditions are promising in absorbing the wave energy when the damping ratio is less than 20% and produces the reflections when damping ratio is more than 20% due to dropping the higher order terms in the expansion. The VABC produces spurious reflections when the waves are not impinging in the normal direction, and reflections increase as the angle between the wave propagation and the normal direction increases. The study extends to provide an efficient absorbing method by combining VABC boundary conditions with Absorbing Layers by Increase in Damping (ALID). The main objective of ALID is to attenuate the reflected waves from VABC in cases of angle incidence. The combination of ALID and VABC, i.e., ALID+VABC, is achieved by matching the impedance of VABC with the last layer of ALID. Results from ALID+VABC are compared with other methods such as ABC, ALID, and SRM (Stiffness Reduction Method). A sensitivity analysis is carried out to verify the efficiency of absorbing the propagation wave energy at various loading frequencies. ALID+VABC has been found to be numerically efficient across a wide range of loading frequencies when compared to the other methods. The method also requires shorter absorbing region lengths, which allows for a smaller number of absorbing layers. However, all the absorbing layer methods such as ALID, ALID+VABC, SRM, and PML are poor at allowing smooth propagation of the wave through layers when waves are entering at a higher incident angle. Dynamic Soil-Structure-Interaction analysis is carried out on a three-dimensional tall building with 20 stories using ABC, ALID, and ALID+VABC as a radiating boundary condition. The complete Soil-Structure-Interaction (SSI) analysis is carried out in two stages. First, a nonlinear static analysis is carried out with gravity loading. The absorbing layers in ALID and ALID+VABC were also present in the static analysis since the damping properties did not influence the analysis. Later, nonlinear dynamic

Abstract

We design a deep learning-based AI algorithm for analysing paediatric patients’ optical coherence tomography (OCT) scans. Our system can distinguish normal from abnormal paediatric retinal OCTs in binary and six-group classification, followed by a ten-layer segmentation of retinal OCT scan and nystagmus detection by performing gaze tracking. Our AI system successfully differentiated normal and abnormal scans with 97.68 % accuracy. Furthermore, the six-point classification system (normal, grade 1-4 FH and atypical FH) achieved a 93.54 % validation accuracy. We have demonstrated the use of AI in classifying paediatric OCT, which can help in the automated diagnosis of abnormal retinal development like Nystagmus at an early age. To extend our work, we introduce segmentation of the ten retinal layers of the central scan. Since individual layer thinning or thickening may be markers of retinal disorders or precursors to future visual loss; hence segmenting individual layers and location-specific measuring their thicknesses is critical in therapeutic practice. By implementing the state-of-the-art DeepLabV3 model, we aim to achieve fully automated segmentation of each of the ten layers of retinal OCT scan. We use the ten carefully segmented ground truth data of the patients, including left and right eyes. We have achieved 0.834 as our mean IoU and 0.896 as our Dice Score on the best-performing DeepLabV3 ResNet-50 semantic segmentation model. Our model accurately predicts all ten layers of the retinal OCT scan, which would be utilized for calculating layer thinning or thickening in later stages of the project. We can also determine regional thickness, brightness, or texture-based indices of individual layers by retinal layer segmentation. This work on layer segmentation contributes to our understanding of retinal or optic nerve head (ONH) disease processes and is used to assess disease status, treatment responses, and visual function, among many others. Our Gaze Tracking model helps with nystagmus waveform generation. Nystagmus is characterised by rhythmic, abnormal eye movements that begin with a ”slow” movement that moves the eye away from the target and ends with a second movement that returns the eye to the target. Horizontal, vertical, torsional, or a mixture of these movements can be present. Hence to tackle such a problem with machine learning, we design an algorithm that could replace the Gold standard Eyelink II that is used to get the waveform for Nystagmus disease. Our method can properly generate such waveforms with high accuracy. This algorithm might get extended to replace the expensive machinery used in current ophthalmology labs. The video data used to test our model has been taken via mobile phones, which further assists in replacing the need for physical presence at the clinic for the diagnosis.

Abstract

Consider an autonomous agent capable of observing multiple humans making a pizza and making one the next time! Motivated to contribute towards creating systems capable of understanding and reasoning instructions at the human level, in this thesis, we tackle procedure learning. Procedure learning involves identifying the key-steps and determining their logical order to perform a task. The first portion of this thesis focuses on the datasets curated for procedure learning. Existing datasets commonly consist of third-person videos for learning the procedure, making the manipulated object small in appearance and often occluded by the actor, leading to significant errors. In contrast, we observe that videos obtained from first-person (egocentric) wearable cameras provide an unobstructed and clear view of the action. To this end, for studying procedure learning from egocentric videos, we propose the EgoProceL dataset. However, procedure learning from egocentric videos is challenging because the camera view undergoes extreme changes due to the wearer’s head motion and introduces unrelated frames. Due to this, current state-of-the-art methods’ assumptions that the actions occur at approximately the same time and are of the same duration do not hold. Instead, we propose to use the signal provided by the temporal correspondences between key-steps across videos. To this end, we present a novel self-supervised Correspond and Cut (CnC) framework that identifies and utilizes the temporal correspondences between the key-steps across multiple videos to learn the procedure. We perform experiments on the benchmark ProceL and CrossTask datasets and achieve state-of-the-art results. In the second portion of the thesis, we look at various approaches to generate the signal for learning the embedding space. Existing approaches use only one or a couple of videos for this purpose. However, we argue that it makes key-steps discovery challenging as the algorithms lack an inter-videos perspective. To this end, we propose an unsupervised Graph-based Procedure Learning (GPL) framework. GPL consists of the novel UnityGraph that represents all the videos of a task as a graph to obtain both intra-video and inter-videos context. Further, to obtain similar embeddings for the same key-steps, the embeddings of UnityGraph are updated in an unsupervised manner using the Node2Vec algorithm. Finally, to identify the key-steps, we cluster the embeddings using KMeans. We test GPL on benchmark ProceL, CrossTask, and EgoProceL datasets and achieve an average improvement of 2% on third-person datasets and 3.6% on EgoProceL over the state-of-the-art. We hope this work motivates future research on procedure learning from egocentric videos. Furthermore, the unsupervised approaches proposed in the thesis will help create scalable systems and drive future research toward creative solutions

Abstract

Cell division refers to the process by which cells grow, replicate their genetic material, and divide to form daughter cells. Major biological processes, namely reproduction, development, wound healing and tissue regeneration, require cell division. Cells switch between quiescence and proliferation states for maintaining tissue homeostasis and regeneration. The cell division process is regulated by a wide range of extracellular and intracellular cues like growth factors, stress, and reactive oxygen species (ROS). The decision to exit or enter quiescence is dysregulated in proliferative and degenerative diseases. Hence, understanding the molecular mechanisms that control the reversible transition between quiescence and proliferation is crucial. In this thesis, we study the regulatory network involved in this decision-making in normal and disease (cancers and Alzheimer’s Disease) conditions and characterize the metabolic adaptation of cancers using systems biology approach. At the restriction point (R-point), cells become irreversibly committed to the completion of the cell cycle independent of mitogen. The mechanism involving hyperphosphorylation of retinoblastoma (Rb) and activation of transcription factor E2F is linked to the R-point passage. However, stress stimuli trigger exit from the cell cycle back to the mitogen-sensitive quiescent state after Rb hyper-phosphorylation, but only until APC/C-Cdh1 inactivation. In the work presented here, we developed a mathematical model to investigate the reversible transition between quiescence and proliferation in mammalian cells with respect to mitogen and stress signals. The model integrates the current mechanistic knowledge and accounts for the recent experimental observations with cells exiting quiescence and proliferating cells. We show that Cyclin E-Cdk2 couples Rb-E2F and APC/C-Cdh1 bistable switches and temporally segregates the Rpoint and the G1/S transition. A redox-dependent mutual antagonism between APC/CCdh1 and its inhibitor Emi1 makes the inactivation of APC/C-Cdh1 bistable. We show that the levels of Cdk inhibitor (CKI) and mitogen control the reversible transition between quiescence and proliferation. Further, we propose that shifting of the mitogeninduced transcriptional program to G2-phase in proliferating cells might result in an intermediate Cdk2 activity at the mitotic exit and the immediate inactivation of APC/CCdh1. Our study builds a coherent framework and generates hypotheses that have been confirmed by experimental findings. Proliferative diseases like cancer arise due to alterations in the regulation of the cell cycle. An emerging hallmark of cancer is metabolic reprogramming, which presents opportunities for cancer diagnosis and treatment based on metabolism. A comprehensive metabolic network analysis of renal cell carcinoma (RCC) subtypes, including clear cell, papillary, and chromophobe, was performed by integrating transcriptome data with the human genome-scale metabolic model to understand the coordination of metabolic pathways in cancer cells. We identified metabolic alterations of each subtype with respect to tumor-adjacent normal samples and compared them to understand the differences between subtypes. We found that genes of amino acid metabolism and redox homeostasis are significantly altered in RCC subtypes. Chromophobe showed metabolic divergence compared to other subtypes with upregulation of genes involved in glutamine anaplerosis and aspartate biosynthesis. A difference in transcriptional regulation involving HIF1A is observed between subtypes. We identified E2F1 and FOXM1 as other major transcriptional activators of metabolic genes in RCC. These results highlight the crosstalk between metabolism and cell division. Further, the co-expression pattern of metabolic genes in each patient showed variations in metabolism within RCC subtypes. We also found that co-expression modules of each subtype have tumor stage-specific behavior, which may have clinical implications. Intriguingly, cell cycle dysregulation triggers not only proliferative diseases such as cancers but also drives degenerative diseases like Alzheimer's disease (AD). Aberrant production and aggregation of amyloid beta oligomers (Aβ) into plaques is a frequent feature of AD. However, therapeutic approaches targeting Aβ accumulation fail to reverse or inhibit disease progression. The approved cholinesterase inhibitor drugs are also mostly symptomatic treatments. During human brain development, the progenitor cells differentiate into neurons and switch to a postmitotic, resting state. However, cell cycle re-entry often precedes the loss of neurons. In this study, we developed mathematical models of multiple routes leading to cell cycle re-entry in neurons that incorporate the crosstalk between cell cycle, neuronal and apoptotic signaling mechanisms. We show that the integration of multiple feedback loops influenc

Abstract

The detection of analytes using Radio Frequency (RF) based sensors has gained immense popularity in recent times. The change in the sensor’s resonant frequency in the presence of different analytes is used as the basis for detection. Although, there have been plethora of advancements in developing RF biosensors, yet most of them have been designed and fabricated considering air all around them. However, in a practical scenario, a solute is mostly used with solvent for testing. The solvent alone produces huge shift in resonant frequency whereas, the shift due to solute and solvent together is significantly narrower (in MHz) when compared to the solvent alone. This indicates that the shift due to solute is narrowband in nature. Most of these sensors utilize traditional Vector Network Analyzers (VNAs) for measuring the s-parameters. However, traditional VNAs are broadband devices that are expensive, large and cumbersome to operate. These VNAs also have much larger bandwidths than the observed frequency shifts as mentioned above and often prove to be overkill. Thus, the optimal solution is to develop narrowband measurement platforms that precisely monitor the frequency shifts. This article tackles the challenge and presents a solvent specific integrated solution for the detection of materials having different permittivities. An Integrated Biosensing Network Analyser (IBNA) is designed and fabricated using microstrip line technology to detect different analytes (Glucose and Sucrose). IBNA primarily consists of a sensing and a measuring unit specifically designed for solvent like water. While the Interdigitated Capacitor based RF Bio Liquid Cavity resonating at 2.3 GHz performs the sensing operation, the Dual Band Six-Port Reflectometer (DBSPR), with an operating frequency of 1.1 GHz and 2.3 GHz, participates in the measurement of the s-parameters. Furthermore, to demonstrate the working of the design, the detection of 1 M Glucose and Sucrose is performed by EM-simulations and experiments. The sensing is carried out efficiently, with Glucose and Sucrose yielding frequency shifts of 15 MHz and 20 MHz at room temperature respectively. The results help in establishing the utility of IBNA and its superiority over traditional network analyzers

Abstract

One of the significant causes of life loss during an earthquake is the collapse of structures in the near-fault regions. Broadly, two sets of parameters affect the structural response, namely: (i) characteristics of structures, like configuration, stiffness, strength, and ductility, and (ii) characteristics of ground motions. Usually, the former are dependent on the latter. But the latter is derived independently by geologists, seismo-tectonists, seismologists, geotechnical engineers and structural engineers. Often, the considered seismic hazard has uncertainties arising from the way strong ground motion is considered in the near-fault regions. To understand and quantify the hazard in the near-fault regions, the ground motions characteristics are examined of the strong motion records of 1999 Chi-Chi, Taiwan earthquake. It is observed that the ground motions in the near-fault region exhibit higher PGA values on the hanging wall side as compared to the footwall side. Further, the behaviours of two reinforced concrete buildings with moment resisting frames of five and ten storeys are studied when subjected to two types of horizontal near-fault ground motions with and without velocity pulse. The responses observed and the damage caused in the moment frame buildings owing to these nearfault ground motions are identified. It is observed that the near-fault ground motions with pulse induce higher drift demands compared to those without pulse and far-fault ground motions. Also, the response of the structure subjected to near-fault ground motion with pulse indicate the impulsive action that requires a revised mitigation strategy other than the conventional earthquake resistant design procedures. Inclusion of a structural wall is identified as a simple solution to control the damage arising from near-fault ground motion effects. The maximum drift demand on the structures with structural walls with SPD of 2.75%, 3.75% and 9% are studied and improved performance is observed as the structural plan density of structural walls is increased.

Abstract

The Internet has many tools that connect people from different ends of the globe. Geographical boundaries do not hinder the connection between two people. Communication is easily facilitated, and news is easily spread. However, this connectivity can be exploited and used negatively. Bullying, hate speech, and fake news can all be spread across the internet, and the common denominator in combating these issues is Sentiment Analysis. Sentiment Analysis is the task of finding the sentiment (positivity or negativity) of a text. This process can be an essential feature for bullying detection, hate speech detection, and fake news detection models. However, there are many libraries and functions available to use, and choosing the optimal one that has the most accuracy to use as a feature for other Natural Language Processing (NLP) tasks is essential. During our search to find functions that calculate sentiment, otherwise known as pre-trained sentiment models, we find that most of the models are trained on polarised texts and datasets that already have sentiment annotated. Hence, we use a nonpolarised text form: news articles and we use a dataset that is not meant to be used with Sentiment Analysis. This helps us test different models. Through our experiments we find that VADER, Stanza, and Transformer’s distilBERT work with a good accuracy, and are ideal to use as the basis for calculating a sentiment feature whereas lexical-based models such as TextBlob and using SentiWordNet are less accurate in its calculations. The dataset we use is the Reuter 50 50 dataset that is meant to be used for another NLP task: Author Identification/Author Profiling. [60] Working with the dataset led to our second task: to create a model that works with minimal requirements for data and computational power but still gives a comparable accuracy to other deep learning methods of Author Identification. Author Identification is that task of being given a random text that could have been written by a group of suspect authors, and finding with reasonable accuracy, which author wrote the text. This is useful to stop other people taking credit for works they have not written and giving credit to the rightful author. We find that using a mixture of word-based features, stylometric feature, and syntax/punctuation features are good features to train a model for Author Identification. We also find that the Boosting class of algorithms outperforms other classes such as Linear models, Nearest Neighbour-Based Models, and Tree Models. XGBoost performs the best out of all the algorithms we test. Linear Models such as SVMs, the Naive Bayes, and the K-Nearest Neighbour algorithm perform abysmally.