ISI Digital Commons (Indian Statistical Institute )
Not a member yet
7571 research outputs found
Sort by
Modulation of plant photosynthetic processes during metal and metalloid stress, and strategies for manipulating photosynthesis-related traits
No full textMetals constitute vital elements for plant metabolism and survival, acting as essential co-factors in cellular processes which are indispensable for plant growth and survival. Excess or deficient provision of metal/metalloids puts plant\u27s life and survival at risk, thus considered a potent stress for plants. Chloroplasts as an organelle with a high metal demand form a pivotal site within the metal homeostasis network. Therefore, the metal-mediated electron transport chain (ETC) in chloroplasts is a primary target site of metal/metalloid-induced stresses. Both excess and deficient availability of metal/metalloids threatens plant\u27s photosynthesis in several ways. Energy demands from the photosynthetic carbon reactions should be in balance with energy output of ETC. Malfunctioning of ETC components as a result of metal/metalloid stress initiates photoinhiition. A feedback inhibition from carbon fixation process also impedes the ETC. Metal stress impairs antioxidant enzyme activity, pigment biosynthesis, and stomatal function. However, genetic manipulations, nutrient management, keeping photostasis, and application of phytohormones are among strategies for coping with metal stress. Consequently, a comprehensive understanding of the underlying mechanisms of metal/metalloid stress, as well as the exploration of potential strategies to mitigate its impact on plants are imperative. This review offers a mechanistic insight into the disruption of photosynthesis regulation by metal/metalloids and highlights adaptive approaches to ameliorate their effects on plants. Focus was made on photostasis, nutrient interactions, phytohormones, and genetic interventions for mitigating metal/metalloid stresses
Multi-Stream Scheduling of Inference Pipelines on Edge Devices - A DRL Approach
No full textLow-power edge devices equipped with Graphics Processing Units (GPUs) are a popular target platform for real-time scheduling of inference pipelines. Such application-architecture combinations are popular in Advanced Driver-assistance Systems for aiding in the real-time decision-making of automotive controllers. However, the real-time throughput sustainable by such inference pipelines is limited by resource constraints of the target edge devices. Modern GPUs, both in edge devices and workstation variants, support the facility of concurrent execution of computation kernels and data transfers using the primitive of streams, also allowing for the assignment of priority to these streams. This opens up the possibility of executing computation layers of inference pipelines within a multi-priority, multi-stream environment on the GPU. However, manually co-scheduling such applications while satisfying their throughput requirement and platform memory budget may require an unmanageable number of profiling runs. In this work, we propose a Deep Reinforcement Learning (DRL)-based method for deciding the start time of various operations in each pipeline layer while optimizing the latency of execution of inference pipelines as well as memory consumption. Experimental results demonstrate the promising efficacy of the proposed DRL approach in comparison with the baseline methods, particularly in terms of real-time performance enhancements, schedulability ratio, and memory savings. We have additionally assessed the effectiveness of the proposed DRL approach using a real-time traffic simulation tool IPG CarMaker
New Techniques to Perform Cross-Validation for Time Series Models
No full textModel validation for time series models has always been a challenge due to a lot of complexities. The presence of auto-correlation in the data creates a challenge to the conventional cross validation techniques like k-fold cross validation to be implemented for time-series models. In this paper, two weighted k-fold time series split cross-validation techniques are proposed for this purpose. The proposed techniques were validated using the opening price data of cryptocurrency. Mean squared error (MSE), Mean absolute error (MAE) and Mean absolute percentage error (MAPE) were the selected metrics to validate the proposed techniques. Both the techniques were found to give robust results; however, the Exponential weighted K-fold time series split cross validation (EWKCV) technique was seen to perform better than Generally weighted K-fold time series split cross validation (GWKCV) technique. The results of the proposed techniques, along with the results of simple train-test split for the time-series models, is seen to give better result
On Exact Feature Screening in Ultrahigh-Dimensional Binary Classification
No full textWe propose a new model-free feature screening method based on energy distances for ultrahigh-dimensional binary classification problems. With a high probability, the proposed method retains only relevant features after discarding all the noise variables. The proposed screening method is also extended to identify pairs of variables that are marginally undetectable but have differences in their joint distributions. Finally, we build a classifier that maintains coherence between the proposed feature selection criteria and discrimination method, and also establish its risk consistency. An extensive numerical study with simulated and real benchmark datasets shows clear and convincing advantages of our proposed method over the state-of-the-art methods. Supplementary materials for this article are available online
ON PRODUCTS OF SYMMETRIES IN VON NEUMANN ALGEBRAS
No full textLet R be a type II1 von Neumann algebra. We show that every unitary in R may be decomposed as the product of six symmetries (that is, self-adjoint unitaries) in R, and every unitary in R with finite spectrum may be decomposed as the product of four symmetries in R. Consequently, the set of products of four symmetries in R is norm-dense in the unitary group of R. Furthermore, we show that the set of products of three symmetries in a von Neumann algebra M is not norm-dense in the unitary group of M. This strengthens a result of Halmos and Kakutani which asserts that the set of products of three symmetries in B(H ), the ring of bounded operators on a Hilbert space H, is not the full unitary group of B(H )
On the directional nature of celestial object\u27s fall on the earth (Part 1: Distribution of fireball shower, meteor fall, and crater on earth\u27s surface)
No full textThis paper investigates the directional distribution of extraterrestrial objects (meteors, fireballs) impacting Earth\u27s surface and forming craters. It also introduces a novel directional statistical mixture model to analyze their falls, validated through rigorous testing. First, we address whether these falls follow non-uniform directional patterns by explicitly employing directional statistical tools for analysing such data. Using projection techniques for longitude and latitude and more importantly, a general spherical statistical approach, we statistically investigate the suitability of the von Mises distribution and its spherical version, the von Mises-Fisher distribution, (a maximum entropy distribution for directional data). Moreover, leveraging extensive data sets encompassing meteor falls, fireball showers, and craters, we propose and validate a novel mixture von Mises-Fisher model for comprehensively analysing extraterrestrial object falls. Our study reveals distinct statistical characteristics across data sets: fireball falls exhibit non-uniformity, while meteor craters suggest a potential for both uniform and von Mises distributions with a preference for the latter after further refinement. Meteor landings deviate from a single-directional maximum entropic distribution; we demonstrate the effectiveness of an optimal 13-component mixture von Mises-Fisher distribution for accurate modelling. Similar analyses resulted in 3- and 6-component partitions for fireball and crater data sets. This research presents valuable insights into the spatial patterns and directional statistical distribution models governing extraterrestrial objects\u27 fall on Earth, useful for various future works
On Weighted Least Squares Estimators for Chirp Like Model
No full textIn this paper we have considered the chirp like model which has been recently introduced, and it has a very close resemblance with a chirp model. We consider the weighted least squares estimators of the parameters of a chirp like model in presence of an additive stationary error, and study their properties. It is observed that although the least squares method seems to be a natural choice to estimate the unknown parameters of a chirp like model, the least squares estimators are very sensitive to the outliers. It is observed that the weighted least squares estimators are quite robust in this respect. The weighted least squares estimators are consistent and they have the same rate of convergence as the least squares estimators. We have further extended the results in case of multicomponent chirp like model. Some simulations have been performed to show the effectiveness of the proposed method. In simulation studies, weighted least squares estimators have been compared with the least absolute deviation estimators which, in general, are known to work well in presence of outliers. One EEG data set has been analyzed and the results are quite satisfactory
Optimal discrimination of quantum sequences
No full textA key concept of quantum information theory is that accessing information encoded in a quantum system requires us to discriminate between several possible states the system could be in. A natural generalization of this problem, namely, quantum sequence discrimination, appears in various quantum information processing tasks, the objective being to determine the state of a finite sequence of quantum states. Since such a sequence is a composite quantum system, the fundamental question is whether an optimal measurement is local, i.e., comprising measurements on the individual members, or collective, i.e., requiring joint measurement(s). In some known instances of this problem, the optimal measurement is local, whereas in others, it is collective. But, so far, a definite prescription based solely on the problem description has been lacking. In this paper, we prove that if the members of a given sequence are secretly and independently drawn from an ensemble or even from different ensembles, the optimum success probability is achievable by fixed local measurements on the individual members of the sequence, and no collective measurement is necessary. This holds for both minimum-error and unambiguous state discrimination paradigms
Oriented total-coloring of oriented graphs
No full textA proper n-coloring of a graph G is an assignment of colors from {1,…,n} to its vertices such that no two adjacent vertices get assigned the same color. The chromatic number of G, denoted by χ(G), refers to the smallest n such that G admits a proper n-coloring. This notion naturally extends to edge-colorings (resp. total-colorings) when edges (resp. both vertices and edges) are to be colored, and this provides other parameters of G: its chromatic index χ′(G) and its total chromatic number χ″(G). These coloring notions are among the most fundamental ones of the graph coloring theory. As such, they gave birth to hundreds of studies dedicated to several of their aspects, including generalizations to more general structures such as oriented graphs. They include notably the notions of oriented n-colorings and oriented n-arc-colorings, which stand as natural extensions of their undirected counterparts, and which have been receiving increasing attention. Our goal is to introduce a missing piece in this line of work, namely the oriented counterparts of proper n-total-colorings and total chromatic number. We first define these notions and show that they share properties and connections with oriented (arc) colorings that are reminiscent of those shared by their undirected counterparts. We then focus on understanding the oriented total chromatic number of particular types of oriented graphs, such as oriented forests, cycles, and some planar graphs. Finally, we establish a full complexity dichotomy for the problem of determining whether an oriented graph is totally k-colorable. Throughout this work, each of our results is compared to what is known regarding the oriented chromatic number and oriented chromatic index. We also disseminate some directions for further research on the oriented total chromatic number