Abstract

Millimeter Wave (mmWave) communication is considered as one of the important frontiers in the suite of technologies for next generation wireless communication, as it offers a huge unlicensed spectrum and can accommodate ever-increasing data traffic and user demands. However, these potential benefits are accompanied by many challenges that include path loss, loss due to blockages, limited coverage, and absorption and penetration losses. Highly directional beamforming and advanced beamforming techniques are considered as potential ways to overcome these losses. The small wavelength of mmWaves allows the packing of a large number of antenna elements in small volumes, thus enabling MIMO implementation and resulting in large beamforming gains from MIMO. Here, we focus on the design of mmWave communication systems to overcome some of the key challenges encountered while exploiting the benefits of MIMO. The high cost and power consumption of mixed signal components at mmWave frequencies make the practical implementation of MIMO beamformers difficult. Hybrid beamforming, which offers a trade-off between analog and fully-digital beamforming, is a potential technique to overcome this challenge. In this thesis, low-complexity hybrid beamforming architecture is adopted in all the proposed mmWave communication system designs in order to achieve practical feasibility. In particular, to address the underlying challenges, we propose the design of hybrid transceivers for MIMO equipped mmWave systems in different communication scenarios: (i) multi-user MIMO downlink communication, (ii) multi-user MIMO interference channel-based communication, and (iii) MIMO amplify-forward (AF) relay-assisted cooperative communication. For all the considered communication scenarios, we propose low-complexity hybrid transceiver designs. To obtain these, first, the fully-digital filters are derived by solving a specifically designed optimization problem, which is then decomposed into RF/baseband hybrid filters using sparse approximation techniques. The distinguishing factor among these scenarios is the difference in the optimization problem considered to address related communication challenges. Specifically, the first scenario considers that the base-station (BS) is simultaneously communicating with multiple user-equipments (UEs) over a MIMO downlink channel. A sum-meansquare-error (SMSE) minimization problem is formulated under a constraint on the total transmit power. This is a non-convex problem and a closed-form solution is hard to achieve. We propose a joint iterative algorithm for a reliable solution towards this problem. The second scenario, which is a multi-user MIMO interference channel-based scenario, considers a complex network, wherein several mmWave transmitters are simultaneously communicating with their intended receivers. The signal from all remaining transmitters is co-channel interference at each intended receiver. Three different designs are obtained for this scenario. First, an overall SMSE is minimized under total transmit power constraint for all the transceiver pairs to obtain the low-complexity mmWave system design. Later, its dual problem, i.e., minimization of transmit power at the transmitting units while achieving the desired quality-ofservice (QoS) criterion (in terms of SMSE), is addressed. The solution to both the SMSE minimization and its dual problem is derived by an iterative algorithm. At last, the maximization of signal-to-leakageplus-noise-ratio (SLNR) under a constraint on the transmit power is considered. The proposed leakagebased problem results in a set of decoupled sub-problems and hence admits a closed-form solution. This leads to reduction in computation complexity as compared to the first two iterative solutions obtained for this scenario. The third scenario considers a MIMO AF relay-assisted communication system design in which two UEs are communicating with each other via an AF relay. This case addresses the short-range and non-line-of-sight (NLOS) communication challenges of mmWaves. Again, an SMSE minimization problem is considered with a constraint on the total relay transmit power. Two design solutions are obtained based on the operating mode viz., half-duplex (HD) mode and in-band full-duplex (FD) mode, of all the communicating nodes. First, a transceiver and relay-filter design are proposed considering the HD AF relay and HD UEs. Further, this design is extended to in-band FD mode where all the nodes operate in FD mode and utilize the same frequency resources. In in-band FD communication, due to simultaneous transmission and reception, the problem of loopback-self-interference (LSI) prevails. The LSI can be mitigated using the existing cancellation techniques. However, these do not provide perfect cancellation and hence some residual LSI still remains. This residual LSI accumulates over time and hugely affects the overall system performance. In the pro

Abstract

With the rise in popularity of social networks, news is no longer monopolized by newspapers, TV and the radio. Most young people get their information about daily events from Twitter’s Trending page or Facebook’s News Updates and other such online forums. Traditional print media no longer remains the preferred way to get access to information anymore. However, this leads us to an unprecedented problem of content verification. While social media platforms have empowered many, by giving voices to people in remote corners of the world, it has also become a double edged sword. Anyone can now tweet or post whatever they want without any factual evidence or proof online. Such posts can gain traction and lead to a lot of misleading information and confusion for an unsuspecting reader. Taking just the case of Twitter, a popular social media website, Twitter sees an average of around 6000 Tweets per second. This translates to nearly 500 million tweets per day. Manually fact checking data on such a scale is an impossible task, and this is just for one such social media platform. Hence we turn towards automated approaches to the problem. While the sheer quantity already poses a huge challenge, this problem is compounded by the problem of the spread of fake news. A study by Sinan Aral, at the MIT Sloan School of Management, showed that falsehood diffused significantly farther, faster, deeper and more broader than the truth in social media platforms like Twitter. This echoes the quote from Mark Twain that ”a lie can travel halfway around the world, while the truth is putting on its shoes.” This makes the problem of automated detection harder as we need not only an accurate system but a fast one as well, as this becomes a time sensitive problem. With the recent advances and breakthroughs in the field of Machine Learning and Deep learning, we have been seeing great results for classification problems. We leverage these innovations to explore their use and effectiveness in the problem of fake news and misinformation. In this thesis, we build a novel architecture for fast and accurate fake news detection in T 3N. T 3N (Text and Temporal Tree Network), leverages deep learning based architectures to encode text, user and tree information in a temporal-aware manner. In Chapter 4, we also look at the identification of fake news spreaders in social media. Empirical observations showed that tweets shared are significant in identifying if the author is a spreader of fake news or other wise and hence we look at identification of fake news spreaders based vi vii on the content shared by them. Chapter 5 studies multi-modal fake news which consist of text and corresponding images and looks at the relationship between images corresponding to a sample and the text. Finally in Chapter 6, we look at the problem in the Indian context. India a country of 122 Major languages and 1599 other languages, is bound see a problem of fake news in regional languages. We try to put forward a first Indian Language dataset for fake news and see if it can be used to help push the frontier for work in regional languages.

Abstract

Recent studies in neuroscience suggest that a unique individual-specific functional connectivity pattern, or brain signature exists in individuals which can be identified by capturing correlated brain activity using functional Magnetic Resonance Imaging (fMRI). All the previous research done in this subject, brief as it is, has utilized some sort of unsupervised similarity measure to identify the individuals, along with mostly using only a static functional connectivity measure. The objective of our thesis is to utilize the functional connectivity patterns to identify an individual using a supervised machine learning based classification approach. To this end, we use the results of static and dynamic functional connectivity (sFC and dFC, respectively) measures on different fMRI datasets as feature sets for different linear and non-linear classification models to evaluate how well the individuals can be identified. For this, we utilize two previously published datasets comprising of naturalistic music listening task-based fMRI response and resting-state fMRI response. We also aimed to find the neural correlates to the brain signatures for identifying the individuals. We found that the classification models using dynamic FC outperform their counterpart, more specifically, the Instantaneous Phase Synchrony (IPS) method of dFC analysis works better than correlation-based windowed analysis. we also found homotopic connectivities of various regions in brain to be an important part of the brain networks used to identify individuals, while the Olfactory Cortex came up repeatedly as an important brain region as well. For the next part of our study, we observed the effects of band-pass filtering and changing the sample size on identifying individuals using the IPS method of dFC analysis by utilizing the minimally preprocessed HCP dataset. We found that performing a band-pass filter greatly boosts the classification accuracy, which stays stable even when the sample size of individuals is increased. Overall, this study shows a lot of promise on using a machine learning approach to identify individuals while using IPS as a dFC measure and presents a novel direction to investigate further

Abstract

Over the years, Earthquake has been one of the leading cause of life loss and destruction of properties. Man-made structures like buildings, roads, dams and bridges are severely affected due to earthquakes. Bhuj earthquake of 2001 killed nearly 15,000 people and damaged around 3.5 lakh buildings. The major contributors to the loss of so many buildings were the poor construction and maintenance practices. To address the deficiencies that lead to such catastrophic damage, major reforms have been introduced to the Indian Standard codes and numerous studies have been conducted to prescribe methods to curtail the damages further. These studies can broadly be classified into earthquake prediction studies and earthquake preparedness studies. Earthquake prediction studies can be used to minimize the loss to life. But over the years, the research on earthquake prediction hasn’t provided satisfactory results, and Earthquake prevention looks like a far-fetched dream. Hence, earthquake preparedness is of utmost importance to minimize earthquake damage. Earthquake preparedness involves the construction of new buildings in compliance with the latest building construction codes of the country as well as fixing the deficiencies of the existing buildings. Hence, identifying the vulnerabilities in a building is a very crucial step towards earthquake risk mitigation. When the vulnerable buildings of a region are identified, the seismic risk of the region can be determined, and actions to mitigate these risks can be taken. But the identification of vulnerable buildings in a region is a very resource and time-intensive process. A quick method to identify vulnerabilities in the buildings of a particular region is Rapid Visual Screening (RVS). This method can be used to find common vulnerabilities in the buildings, and the entire process of RVS of a building can be completed within 30 minutes by a single, trained person. But many buildings may require a detailed investigation in order to assess its safety. Detailed investigation requires the modelling and analysis of the building in finite element software. This requires the time and energy of people who are proficient in such software along with powerful machines to run such software. These problems make detailed investigation, time and resource-intensive process as well as very expensive. Hence, there is a need for an efficient and effective method to perform this task.

Abstract

The problem of multi-objective decision making has been an important one in the field of database management systems. Establishing dominance relationships among different data points has useful applications in industries such as tourism & sports. Given a dataset D with c points, a point pi dominates another point pj if it is better than or equal to pj in all the dimensions in space S and better than pj in at least one dimension. The set of such points which dominate all other data points in the dataset is referred to as the skyline set of points. The skyline operator suffers from the curse of dimensionality which means as the number of dimensions increase the number of skyline points become large and difficult to compute. With data growing at exponential rate in today's information age traditional algorithms designed for skyline query computation over limited resource standalone computers will not fare too well. Scaling up systems in terms of computing power & memory has traditionally proven to be expensive. Therefore, there is a need to leverage a cluster of cheap commodity hardware to efficiently compute skylines of large datasets. Thus the data needs to be divided into multiple smaller datasets which can be used to compute individual skylines across the smaller datasets and merged to form the overall skyline of the entire dataset by using a divide and conquer approach. The curse of dimensionality is even more amplified on large datasets hence making the problem of computing skyline even more complex, thus motivating us to use techniques which do not require us to eliminate points already in the computed skyline by points encountered afterwards. We achieve this by approaching points along the Z-order curve which provides us a monotonic access order of data points thus identifying skyline points earlier. The Z-address of a point in multi-dimensions is simply calculated by interleaving the binary representations of its coordinate values. In this thesis, we address the problem of efficient computation of skyline for large datasets using the Map-Reduce programming paradigm. Our approach is to map multi dimensional data to one dimension while preserving the locality of the data points using Z-order curve and exploiting the sort phase of a typical Map-Reduce job to receive these points in the Reduce job automatically sorted in the requisite order. Our results show a reduction of up to 31% in execution time for skyline computation over a Map-Reduce framework using our algorithm as compared to existing methods.

Abstract

Automated Question Answering (AQA) is an attractive variation in search engines where the QA system automatically returns a passage, which answers the user’s question instead of providing several links and letting the user explore for an answer. These AQA systems are an integral part of search engines and voice assistants such as Amazon Alexa, Apple Siri, Microsoft Cortona, Google Assistant, etc., Due to the recent growth in AQA systems, the research in this field has seen a tremendous improvement. Ranking the passages is an important step in an AQA system, where the candidate passages are identified for a question and scored as likely to contain an answer. To explore the various practical approaches for this problem, Microsoft has released a large scale dataset called as Microsoft MAchine Reading & COmprehension (MS MARCO) in 2016. Using the subset of the MS MARCO dataset, it has also organized a challenge called Microsoft AI Challenge India (MSAIC) to provide a platform for researchers and AI practitioners across the country to contribute to this area. In this thesis, we propose different approaches based on biLSTM that we used during our participation in Microsoft AI Challenge India. We show that our best model, based on the biLSTM encoder with coattention mechanism (which we call as coattention encoder), has significantly outperformed the shared baselines and benchmark models. In the second part of this thesis, we discuss and address the two limitations of the coattention encoder. We first extend the coattention encoder to n-grams to make it capture the local context. We further improve the coattention encoder by proposing a simple attention pooling using the query to refine the passage’s representation. We perform all our methods on the MS MARCO passage ranking task and demonstrate that these two methods yields a relative increase of 8.04% in Mean Reciprocal Rank (MRR@10) compared to the naive coattention encoder. It is the best non-transformer model and is competitive to the BERT base while only having fewer parameters. Additionally, when compared to BERT, our model has a very low ranking latencies.

Abstract

Blockchain technology has been one of the most disruptive innovations in the last decade. Blockchain is a decentralized, append-only trustworthy distributed database. It was invented to build Bitcoin – to democratize financial systems, in particular, currency. However, soon the researchers learned the power of blockchain to build smart contracts, programs that can execute and validate contracts, which otherwise were traditionally done through legal contracts. Since then, there are a plethora of applications has been envisioned using blockchain. At the backend, miners are the ones who maintain the blockchain in anticipation of fair rewards. Bitcoin offers two types of rewards for the miners: (i) block rewards that halve every four years and (ii) transaction fees. Halving block rewards is done by making currency inflation-free. Transaction fees are optional. This can lead to stagnation of transactions and unprofitability for miners. Another challenge with Bitcoin or many other blockchains, to build real-world applications, is performance. Many times, improving performance in blockchain leads to different security issues. In this thesis, we address both research challenges: (C1), how to configure the Bitcoin reward structure to maintain its ecosystem healthily as well as inflation-free. (C2), how to design a blockchain that is performant as well as secure. Towards C1, we propose BitcoinF, a simple modification to Bitcoin rewards that ensures fairness to the miners. We analyze it game theoretically. Towards C2, we propose QuickSync, a novel blockchain protocol that is faster than most of the secure blockchain protocols such as Bitcoin and Ourborous and retains all the security properties. We do security analysis for different attacks such as Sybil attacks and prove the robustness of QuickSync towards the attacks we studied

Abstract

Machine learning methods have found applications in various branches of healthcare including disease diagnosis, disease progression and outcome prediction, drug design, etc. The computational nature of such methods democratizes medical expertise by enabling access to diagnosis and outcome prediction in the absence of a medical expert. In this thesis we present two different applications of AI in health care. We present LPGNet an efficient deep learning based model for diagnosis of Parkinson’s disease from Gait. Next, We present the statistical analysis and models built for severity and mortality prediction in hospitalized Indian COVID-19 patients. In the first part of the thesis we present LPGNet an efficient method for diagnosis of Parkinson’s Disease (PD) from gait. PD is a chronic and progressive neurological disorder that results in rigidity, tremors and postural instability. There is no definite medical test to diagnose PD and diagnosis is mostly a clinical exercise. Although guidelines exist, about 10-30% of the patients are wrongly diagnosed with PD. Hence, there is a need for an accurate, unbiased and fast method for diagnosis. In this study, we propose LPGNet, a fast and accurate method to diagnose PD from gait. LPGNet uses Linear Prediction Residuals (LPR) to extract discriminating patterns from gait recordings and then uses a 1D convolution neural network with depth-wise separable convolutions to perform diagnosis. We also undertake an analysis of various cross-validation strategies used in literature in PD diagnosis from gait and find that most methods are affected by some form of data leakage between various folds which leads to unnecessarily large models and inflated performance due to overfitting. The analysis clears the path for future works in correctly evaluating their methods. LPGNet achieves an AUC of 0.91 with a 21 times speedup and about 99% lesser parameters in the model compared to the state of the art which would enable the method to be implemented in embedded systems to enable automated Parkinson’s diagnosis to be cheap and widely accessible. In the second part of the thesis we present the statistical analysis to understand and build models for severity and mortality prediction in hospitalized Indian COVID-19 patients. The clinical course of coronavirus disease 2019 (COVID-19) infection is highly variable with the vast majority recovering uneventfully but a small fraction progressing to severe disease and death. Appropriate and timely supportive care can reduce mortality and it is critical to evolve better patient risk stratification based on simple clinical data, so as to perform effective triage during strains on the healthcare infrastructure. We analyse a diverse set of 70 bio-markers including physiological, haematological and genomic parameters from 544 COVID-19 patients from New Delhi, India to identify significant markers and build risk stratification and mortality prediction models. An XGboost classifier yielded the best performance on risk stratification (AUC score of 0.83). A logistic regression model yielded the best performance on mortality prediction (AUC score of 0.92) when tested on a holdout test set of 115 patients. Examination of the data in comparison to a similar dataset with a Wuhan cohort of 375 patients was undertaken to understand the much lower mortality rates in India and the possible reasons thereof. The comparison indicated higher survival rate in the Delhi cohort even when patients had similar parameters as the Wuhan patients who died. Our analysis of medications administered points to usage of steroids early on which may have helped in the survival of severe patients. This study helps in identifying high-risk patients and provides insights into the low mortality rate in India

Abstract

Use of Immersive Virtual Reality(VR) stack is expanding, as the organizations explore alternative methods to reduce the cost of training. The effectiveness of VR simulators has demonstrated significant gains in the transfer of training for domains such as surgical training, flight training, military reconnaissance, search and rescue operations, etc. However, despite significant advancements, there exists a substantial gap to study human cognition and behavior and to also train them for demanding tasks like military combat and reconnaissance. Many of the military systems in use today are robotic with a limited degree of decision-making ability and interaction with the environment where making mistakes isn’t an option and therefore requires human intervention to operate and command remotely. Thus, teleoperators need to be instructed and trained prior to perform such operations efficiently and effectively. For a better understanding, and increasing VR’s usability, we aim to identify the relevant constituent components of immersive technologies, and differentiate their roles. In this thesis, we aim to determine the role of two determinants (goal-specific instructions and display fidelity) of spatial knowledge that one acquires during spatial navigation in VR. We conducted two experiments to disentangle the contribution of the two factors individually. Literature has demonstrated the relative importance of instructions on task performance and having goals can help in focusing attention on relevant information in the environment. However, it is also dependent on the nature of goal or activity that one is asked to accomplish. From a navigation perspective, the simplest of the goal is to return to the starting point to end the activity. In the first study, we included this goal in the instructions for one group while the other group were only instructed to explore the virtual environment and build understanding. Acquired topological spatial knowledge was assessed from the drawn sketch-maps. Our results showed that participants with goal-specific instructions found it difficult to recall one out of three spatial relationships (Object Positioning). These results indicate that the presence of additional goals in the instruction can impact the spatial memory of the operators and can be detrimental to task performance. HMD VR compared to DT, provides a 3-D experience during the virtual exploration, it offers a sense of complete immersion in the virtual world but literature has shown mixed findings depending on the applications. To study the role of the second factor, we conducted a pilot study followed by a full-scale study to examine how two common VR display modes, head-mounted display (HMD) and desktop (DT), would affect spatial learning when we restrict ambulatory locomotion in HMD. The findings and shortcomings of the pilot study helped in the refinement of experimental design, protocols, and methodology. The two display modes were compared on spatial learning, time taken to complete the virtual exploration, motion sickness, workload, sense of presence, and spatial orientation ability. Our results highlight the differential effect of DT vs. HMD VR depends on the nature of cognitive and functional behavior that the user employs. We explain these results in terms of deficient idiothetic information in non-ambulatory HMD and lesser sensory conflicts in desktop mode. This thesis sets up the foundation stones and provides future research directions that could be used in developing more robust VR experiments and applications. It is intended to be useful for VR experimenters as well as end-users. These results are relevant for developing customized and sustainable virtual reality-based human-computer interactions.

Abstract

Online learning problems are among the most abundant machine learning problems, such as advertisement selection, movie recommendation, and node or link prediction in evolving networks. In online learning, one must make online, real-time decisions and continuously improve performance with the sequential arrival of data. The multi-armed bandit (MAB) problem is one of the fundamental online learning problems that captures the classic exploration vs. exploitation dilemma. As per the applications’ requirements, new variants of MAB have been developed to tackle the problem. For example, strategic bidding in the online learning environment (MAB mechanism); availability of extra information to make the decision (Contextual MAB); selecting multiple arms instead of the single-arm (Combinatorial MAB); a varying set of available arms (Sleeping MAB), etc. We identify some of the problems and appropriately address them by proposing efficient algorithms. We provide theoretical guarantees for the performance of the proposed algorithms and showcase their efficacy through simulations. Following are the problems we address in this thesis: • Designing truthful contextual multi-armed bandits with the backdrop of sponsored search auction [3]. We tackle the drawbacks of the current state-of-the-art. • Propose an efficient algorithm for sleeping combinatorial multi-arm bandit problem when we consider general reward function (satisfying mild smoothness conditions) [2]. • We take a multi-arm bandit approach to subset selection under constraints in an unknown environment [46].