Abstract

Evapotranspiration (ET) is one of the prominent hydrologic variables affecting water and energy balances and critical factors for crop water requirements and irrigation scheduling. ET is a complex hydrological variable defined by various hydro climatological variables. The major forms of ET which are widely applicable in hydrological, water balance, drought assessments and other ecological assessments are Potential (PET) and Actual Evapotranspiration (AET). PET represents the atmospheric water demand with a focus on climatological variables. AET is influenced by climate, vegetation, soil moisture, and the amount of available water and presents the annual water balance between precipitation and latent heat exchange. Various empirical formulations have been developed to estimate PET and AET depending upon the availability of hydrometeorological variables. These empirical formulations are region-specific and developed for particular climatic conditions. Furthermore, the empirical family of models has a major limitation as they require a large number of hydrometeorological inputs, limiting their utility in data-scarce areas of ungauged basins. In this context, empirical and mathematical models have emerged as simple and readily implementable for estimating PET and AET with measured hydrometeorological parameters as independent variables. Such mathematical models can be valuable to predict PET and AET when climate data is insufficient. The present study compared various empirical models and data-driven algorithms to predict PET and AET using various hydroclimatic variables. Four empirical methods, such as FAO-based Penman-Monteith method, temperature-based Hargreaves method, and radiation-based Turc, and Priestley-Taylor method, were used to estimate PET at a daily time scale. Five data-driven algorithms, such as Long short-term memory neural networks (LSTM), Artificial Neural Network (ANN), Gradient Boosting Regressor (GBR), Random Forest (RF) and Support Vector Regression (SVR), were implemented. Two empirical AET models, such as Budyko and Turc methods, were utilized in estimating AET. These models were evaluated over two different climatic regions, Hyderabad, the largest city of the Indian state, Telangana and aipara in New Zealand, both with semiarid climates. Dataset consists of daily meteorological data of maximum and minimum air temperature, relative humidity, solar radiation, and wind speed over a period of 51 years (1965 - 2015) for Hyderabad and for a period of 6 years (2010- 2016) for Waipara station.The Penman-Monteith method was considered as the standard method to compare the different models and various empirical models of PET. The models were trained and tested with climate variables as input variables and various empirical models as reference models. The most influencing climate variables on PET were found in the order of temperature, solar radiation, wind speed, and relative humidity, which formed as the basis to choose different datasets to train over models and compare the results to validate. Temperature and radiation-based models of Turc and Priestley-Taylor methods can be used to estimate PET when all other climate variables are not available as they are also promising with the Penman-Monteith method. The results indicated that 99% accuracy could be achieved with all climatic input, whereas accuracy drops to 86% with limited data. Both LSTM and ANN models have been noted as the most robust models for estimating PET with minimal climate data. Even though the excellent performance can be achievable when all input variables are used, the study, however, found that even a three-parameter combination (temperature, wind speed and relative humidity values) or two-parameter combination (temperature and relative humidity, temperature and wind speed) can also be promising in PET estimation for a semi-arid climate. The AET over semi-arid climatic conditions of Hyderabad, Telangana, India and Waipara (New Zealand) was estimated using modelled and different empirical methods-based PET using Budyko and Turc models. The proposed empirical-based AET models, Budyko and Turc, showed that the AET process has the potential to be estimated by structurally simple methods. Equation-based AET methods made it possible to extract useful information about the hydrological process. It was observed that the meteorological variables of temperature and solar radiation have more significant contributions than other variables in the estimation of AET. In addition, the effects of the meteorological variables were found to be essential and effective in the estimation of AET. The research findings of the study reveal that under limited data availability, the best input combinations were identified as temperature and wind speed for estimating PET; temperature, wind speed and precipitation for estimating AET for semiarid climatology. Overall, the research findings of the study stress on the use of limited data in

Abstract

Quantum computing hardware has progressed rapidly. Simultaneously, there has been a proliferation of programming languages and program optimization tools for quantum computing. Existing quantum compilers use intermediate representations (IRs) where quantum programs are described as circuits. Such IRs fail to leverage existing work on compiler optimizations. In such IRs, it is non-trivial to statically check for physical constraints such as the no-cloning theorem, which states that qubits cannot be copied. We introduce QSSA, a novel quantum IR based on static single assignment (SSA) that enables decades of research in compiler optimizations to be applied to quantum compilation.QSSA models quantum operations as being side-effect-free. The inputs and outputs of the operation are in one-to-one correspondence; qubits cannot be created or destroyed. As a result, our IR supports a static analysis pass that verifies no-cloning at compile-time. The quantum circuit is fully encoded within the def-use chain of the IR, allowing us to leverage existing optimization passes on SSA representations such as redundancy elimination and dead-code elimination. Running our QSSA-based compiler on the QASMBench and IBM Quantum Challenge datasets, we show that our optimizations perform comparably to IBM’s Qiskit quantum compiler infrastructure. QSSA allows us to represent, analyze, and transform quantum programs using the robust theory of SSA representations, bringing quantum compilation into the realm of well-understood theory and practice.

Abstract

The goal of rendering in computer graphics is the synthesis of photorealistic images from abstract virtual scene descriptions. Among the many rendering methods, path tracing along with physically accurate material modeling has emerged as the industry standard for photorealistic image synthesis. Path tracing is, however a computationally expensive operation. Moreover, physically accurate material modeling requires iterative parameter tuning and visualization using path tracing. This quickly becomes a bottleneck for the tuning of a large number of parameters found in most physically accurate material models, and artists end up spending a lot of time waiting for path tracing to converge. This dissertation proposes to leverage the learning power of neural networks to plausibly approximate path tracing for: (a) quick material visualization while editing, (b) out-of-the box material recovery using inverse rendering. The traditional workflow for tuning material parameters and visualization is typically followed by modifications to lighting for better understanding the behaviour of the material. We thus incorporate the ability to modify the environment lighting in our neural material visualization framework. This significantly aids in understanding the behaviour of the material thus aiding the process of parameter tuning, which we demonstrate with a user study. Most real-world materials are spatially varying, which is that their behaviour varies across the surface of the geometry. While previous works in this area dealt with uniform materials, we propose improvements for handling spatially varying materials. We design our neural network architecture to be light-weight, having 10× lesser trainable parameters than the state-of-the-art, while also producing better visualizations. We also build a tool to demonstrate real-time parameter tuning and visualization of materials. Finally, we extend the above work to handle general 3D scenes in specific lighting scenarios. This is done by training a small fully connected neural network on a simple scene in the local tangent frame. Training in the local tangent frame decouples geometric complexity from training which implies that the network generalizes to any geometry. We embed this network in a standard path tracer to evaluate direct illumination at each path vertex. We thus not only obtain noise free direct illumination but also out-of-the box gradients of the rendering process, which can be used for inverse rendering. We show that these gradients can be used for plausibly recovering the material of a target scene. We conclude by discussing various practical aspects of our method.

Abstract

Recent advances in networking sites, the Internet revolution has changed everything, right from businesses to healthcare to education to the ways of communicating with our friends. The Internet has opened doors for people to express themselves, write their thoughts about a particular topic, share an experience online for other people to read without much hassle. Even before going out, order something, or watching a show/movie, people tend to check the online reviews first. And it is feasible and practical also because anyone wouldn’t want to spend their time, money, and other resources on something that isn’t worth it. With so many reviews, tweets, and content available online, it is also important to process them in such a way that they can be used by everyone productively. One use-case could be as the word limit for online reviews on sites like IMDb, Zomato, and Amazon is pretty significant (10, 000 characters for IMDb, Amazon has a limit of 5, 000 words ∼ 23, 500 characters) some reviews tend to be longer. While the reviewer elaborates on their experience, from a reader point of view, what really important is, what aspects of the given target entities the reviewer liked/disliked. This is where the need for Aspect-based Sentiment Analysis arises. Aspect-Based Sentiment Analysis (ABSA) aims to extract the aspects of the given target entities and their respective sentiments. The main issue is, the amount of data is so huge that manually processing the data on such a vast scale is impossible. E.g., Twitter alone sees an average of 6000 tweets per second, roughly 500 million tweets per day. Hence, we look for some fast, automated methods that can do the processing almost in real-time. In this thesis, we propose a deep-learning enabled model, TEASER, based on an extract-then-classify framework for extracting the aspects and detecting the respective sentiment attached. We also conduct extensive experiments on the 3 existing datasets (Restaurant14, Restaurant15, Laptop14) to show that TEASER performs better than the existing models. In chapter 4, we present two novel datasets in the domain of movie reviews, Movie20, and moviesLarge. Movie20 is a supervised dataset of 1162 sentences manually annotated by two human annotators, whereas moviesLarge is a pseudo-labeled dataset of 14373 sentences. With the help of Semi-supervised learning, we benchmark TEASER on the Movie20 dataset. We also evaluate the results of TEASER on the Movie20 dataset thoroughly and try to reason the gaps in predicted output and gold annotation.

Abstract

Pattern mining is an important task of data mining and involves the extraction of interesting associations from large databases. In the literature, pattern mining approaches have been investigated to extract the different kinds of patterns such as frequent patterns, sequential patterns, periodic patterns, utility patterns and coverage patterns from transactional databases. The pattern mining approaches have been widely employed to improve the performance of applications in the areas of recommendation systems, e-commerce, customer relation management, internet advertising, decision support systems, social media and so on. One of the key issues of developing pattern mining algorithms is as follows: a pattern mining algorithm has to process a huge volume of transactional dataset. Research efforts are going on to develop pattern mining algorithms to operate on large datasets by exploiting different parallel processing paradigms, namely shared memory, shared disk, shared-nothing, GPUs, incremental, MapReduce and Grid. Typically, a given transactional database D gets updated due to the addition and/or deletion of transactions. Consequently, some of the previously discovered patterns may become invalid, while some new patterns may emerge. To extract the new knowledge we should mine the database every time it gets updated. One of the main issues is to mine huge Dynamic Databases for every time instant from scratch is computationally costly. The concept of incremental mining is evolved to solve the issue of mining patterns for every instant efficiently for Dynamic Databases and has become an active research area. When the database is updated, with Incremental mining, there is an opportunity to improve the performance by limiting the computation to updated parts of the database and with the required additional processing. In the literature, Incremental mining approaches have been investigated to extract different kinds of patterns such as frequent patterns, sequential patterns, periodic patterns and utility patterns from transactional databases. In this thesis, we propose an improved approach for extracting Coverage Patterns for Dynamic Databases under the incremental paradigm. Given a transactional database, the coverage pattern mining involves the extraction of multiple sets of items, where each set (covers a certain percentage of transactions) satisfies relative frequency, coverage support and the overlap ratio constraints. The applicability of coverage patterns has been demonstrated in improving the performance of search engine advertising, banner advertising and visibility mining. In the literature, a level-wise iterative coverage pattern mining (CPM) and a pattern growth approach have been proposed.In this thesis, we propose a Comprehensive Coverage Pattern Mining (CCPM) approach, for efficiently extracting Coverage Patterns under the incremental paradigm, when a set of transactions are added only, set of transactions are deleted only and both set of transactions are added and deleted from the original database. For each case, we have identified a set of rules by identifying the potential opportunities to avoid unnecessary processing as compared to the rudimentary option of the processing from the scratch. Furthermore, it has been observed that, when a set of transactions are added and/or deleted from the original database, if the order of items in the frequency list (Flist) of the database is changed, additional patterns have to be validated and require additional processing. Based on the rules and by considering the issue of change in the order of items in the Flist, we have proposed improved approaches to extract coverage patterns, by considering the case of (i) addition of transactions (ii) deletion of transactions and (iii) addition and deletion of transactions. We have also performed extensive experiments on two real click-stream datasets and one synthetic dataset and demonstrated the overall effectiveness of our proposed approach.

Abstract

The recent advances in Machine Learning and other data driven techniques have transformed domains like robotics, computer vision and natural language processing. These algorithms are now also being applied to a wide array of chemistry problems as well. One such class of machine learning algorithms is deep generative modelling which has found its application in de novo molecule generation. Generating lead molecules in silico, helps reduce the wastage of limited computational and laboratory resources by reducing the number of generated bad leads. For every task, the lead molecules may exist in different regions of the chemical space which would have to be explored appropriately. The drug discovery process is long, arduous and expensive while also being plagued by the problems caused by bad leads especially for novel targets and the design of new inhibitors for novel targets is now more important than ever especially in the current scenario with the world being plagued by COVID19. Drug molecules must possess certain properties to be effective so that they can be considered for further validation. Conventional computational approaches such as high-throughput virtual screening require extensive combing through existing database and evaluating all the molecules in them in the hope of finding possible hits that possess the required properties. However, existing libraries of druglike molecules, advances in deep generating modelling and optimization techniques like reinforcement learning has provided a new avenue to tackle this problem. In this thesis, a computational strategy is proposed for the de novo generation of molecules with high binding affinities to the specified target and other desirable properties for drug-like molecules using reinforcement learning. A deep generative model built using a stack-augmented recurrent neural network initially trained to generate drug-like molecules is optimized using reinforcement learning to start generating molecules with desirable properties like LogP, Quantitative Estimate of Drug Likeliness,Topological Polar Surface Area, and Hydration Free Energy along with the binding affinity. For multiobjective optimization, a novel strategy was also devised in which the property being used to calculate the reward is changed periodically. In comparison to the conventional approach of taking a weighted sum of all rewards, this strategy shows an enhanced ability to generate a significantly higher number of molecules with desirable properties while also exploring new regions of the chemical space for multiple targets.

Abstract

Radio frequency (RF) sensing is emerging as an alternative for quantitative analysis of bio-samples due to its advantages over conventional techniques. It is robust, reusable, portable, less cumbersome and the quantity of sample is typically in micro-litres. Planar resonant sensors like complementary split ring resonators (CSRR) and inter-digitated capacitors (IDC) are the most excellent choices for biosample detection due to simple design, low cost fabrication, compactness and effortless integration with other components for lab on chip development for various RF applications. CSRR based RF sensor has been designed of size 38 x 32 mm2 is designed with resonant frequency at 1.51 GHz. The designed sensor is fabricated on a 1.6 mm thick FR4 substrate with copper of thickness 0.035 mm on both sides. Quantitative analysis of sucrose, glucose, fructose and L-Lysine have been performed using Keysight Vector Network Analyzer (VNA). Although this sensor is advantageous in various aspects, existing sensors still suffer from a lack of sensitivity in detecting minuscule changes in the dielectric constant. The major work presents the optimization of RF cavity design using binary particle swarm optimization (BPSO) for quantitative analysis of L-Lysine, extendable to any bio-sample. The design of the sensor is broken into fragments called cells. A combination of the cells constitutes a cell pattern resulting in a RF design. The cell pattern is optimized in such a way using BPSO to achieve maximum sensitivity for a particular set of dimensions of the sensing region. Usually, sensors are designed in the presence of air which has a dielectric constant of ∼1. In contrast, most bio-samples have solvents associated with them. Water and its based solvents having dielectric constant ∼79 are most commonly used for bio-sample detection. Therefore, we perform optimization of RF designs like CSRR and IDC for achieving enhanced sensitivity in the presence of the solvent above sensing region, unlike air-based reference. Detecting small changes in the concentration of the sample under test (SUT), which contributed to a minuscule change in the dielectric constant of the solution, has been achieved through optimized RF sensors. The final sensors have been fabricated using the low-cost PCB technology on a FR-4 substrate of dielectric constant ∼4.3. An enhancement of 47.38× for the CSRR and 2.03× for the IDC in the detection of L-Lysine at frequencies 2.18 GHz and 0.72 GHz, respectively has been achieved when compared to sensitivities of their traditional counterparts. Moreover, the sensors have been tested to detect sucrose and glucose, and promising results have been obtained. This methodology, called solvent based optimization, can be extended for enhancement of the sensitivity of any planar RF sensor, making it a suitable contender for optimization of next-gen RF bio-sensors. This work shows great potential in various biomedical, agricultural, and pharmaceutical applications

Abstract

Allostery is the process of regulation of a protein caused by the binding of a ligand at a distant region from the active site. It is one of the most widely studied properties of proteins and biological systems. Ligand binding at allosteric sites lead to a conformational change in the protein, propagated via changes in protein dynamics that alter the capability of the protein to bind to ligands or molecules at active sites. Allostery is required by biological systems to carry out several crucial functions like hormone binding to receptors, protein regulation to name a few. While traditional view of allostery considers large conformational and structural changes, the concept of dynamic allostery challenges this assumption, wherein ligandinduced changes in protein dynamics could produce allosteric communication between distinct binding sites, even in the absence of macromolecular conformational changes. In this thesis, we have considered PDZ domain of protein PSD-95 as our model system to understand the mechanism of dynamic allostery and identify residues involved in allostery. PDZ domains are evolutionary conserved protein-protein interaction modules associated with cellular signaling and are implicated in the localization of membrane receptors and ion channels. The analysis involves using two popular approaches, Molecular dynamics simulations and network-based approach. We show that the combined MD and networkbased approaches together provide a better insight on possible allosteric communication pathways on ligand binding and also is able to improve the prediction of allosteric sites as compared to either of the methods individually. Additionally, in the extension of the webserver, NAPS, we developed interaction energy networks for the analysis of protein structures and support the analysis of MD trajectories such as construct and compare network-based features of protein structures from intermediate time-steps of the MD simulation.

Abstract

Always-on agents are those agents which are like daemon programs that are always on and can do tasks as and when they arrive. These agents are idle when there is no task assigned to them. Further, those agents that work on a task also wait for some event or task completion, and hence, are also idle for short durations while executing the tasks. The question arises what should the agent be doing when it is idle. In this thesis, we conduct an empirical analysis to show improved decision-making capabilities by agents when they exploit their idle time. We analyze scenarios where (i) the agents analyze their past tasks regarding decisions taken and their impact and where (ii) the agents cooperate with other agents to improve their decision-making. Performance improvement can be measured by considering an increase in success rate, avoiding strategies that may not work, quicker decision making, etc. While executing a task, our always-on agent stores pertinent details of the task done in a database, such as decisions taken, paths of execution of the task, goodness of them, etc. The agent uses different strategies to use this stored knowledge. We present and evaluate three strategies that always-on agents can use (i) Frequent Decision Strategy (FDS) - the agent stores the prior executions and their frequency of success and failure, repeats the most frequent successful decision taken during prior executions of the task, (ii) Analyzed Decision Strategy (ADS) - the agent analyses prior executions that were successful or not, stores in the database the goodness of various alternatives and chooses the best alternative and (iii) Online Analysis Decision Strategy (OADS) - the always-on agent while executing its task, during its idle time analyses the possible future situations and prepares the list of best possible decisions that can be taken in future. The FDS and ADS are used when the agent is not doing any task and is off-line. In contrast, OADS generates new mock tasks to consider possible alternative task execution situations to expand its decision-making scenarios while having a task at hand. We conduct our empirical study on always-on agents playing Connect-4 games to check the viability and improved decision-making. We also present idle time analysis in always-on cooperative agents. We propose a Multi-agent Framework that always-on cooperative agents can use in their idle time to improve decision-making. We formulate scenarios in which the agent has a rationale to cooperate with other agents (i) Exception Handling - the agent cooperates to handle situations in which it does not know what to do, (ii) Confidence Boosting - the agent cooperates to boost confidence in actions it wants to take, (iii) To obtain Different Experiences - the agent cooperates to obtain different experiences. We conduct an empirical study on always-on cooperative agents using the Multi-agent Framework and show improved decision-making.

Abstract

Spontaneous speech is a particular type of speech setting where a speaker speaks without preparing in advance. This makes the speaker think about what to say on the spot, formulate the utterances and then produce the speech. Such a setting often leads to abrupt breaks or discontinuities in the normal conversation flow called disfluencies. Disfluencies can provide information regarding the speaking style, speaker identity and language fluency, which can be useful for several speech-based applications. For automatic speech recognition (ASR) systems, the presence of these disfluencies leads to a higher word error rate, since most ASR systems are developed on non-spontaneous read speech data. Thus, the detection of disfluencies in spontaneous speech becomes an essential task for many applications. For training any machine learning system, one need data. In this thesis, we introduce IIITH-IED dataset for disfluencies in spontaneous lecture-mode speech, and then use it to develop frame-level automatic disfluency detection systems. Finally, we propose a transfer learning method to detect disfluencies in spontaneous lecture mode speech using three frame-level automatic disfluency detection systems trained on stuttered speech. Compared to baseline systems, proposed method gives an average improvement of 2.25% across all disfluencies and all detection systems.